Notebook

Running Tune experiments with HEBOSearch¶

In this tutorial we introduce HEBO, while running a simple Ray Tune experiment. Tune’s Search Algorithms integrate with ZOOpt and, as a result, allow you to seamlessly scale up a HEBO optimization process - without sacrificing performance.

Heteroscadastic Evolutionary Bayesian Optimization (HEBO) does not rely on the gradient of the objective function, but instead, learns from samples of the search space. It is suitable for optimizing functions that are nondifferentiable, with many local minima, or even unknown but only testable. This necessarily makes the algorithm belong to the domain of "derivative-free optimization" and "black-box optimization".

In this example we minimize a simple objective to briefly demonstrate the usage of HEBO with Ray Tune via HEBOSearch. It's useful to keep in mind that despite the emphasis on machine learning experiments, Ray Tune optimizes any implicit or explicit objective. Here we assume zoopt==0.4.1 library is installed. To learn more, please refer to the HEBO website.

In [1]:

# !pip install ray[tune]
!pip install HEBO==0.3.2

Collecting HEBO==0.3.2
  Downloading HEBO-0.3.2-py3-none-any.whl (106 kB)
     ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 106.3/106.3 kB 1.6 MB/s eta 0:00:00a 0:00:01
Requirement already satisfied: gpytorch>=1.4.0 in /Users/kai/.pyenv/versions/3.7.7/lib/python3.7/site-packages (from HEBO==0.3.2) (1.6.0)
Requirement already satisfied: scikit-learn>=0.22 in /Users/kai/.pyenv/versions/3.7.7/lib/python3.7/site-packages (from HEBO==0.3.2) (0.24.2)
Collecting GPy>=1.9.9
  Using cached GPy-1.10.0-cp37-cp37m-macosx_10_9_x86_64.whl (1.5 MB)
Requirement already satisfied: torch>=1.9.0 in /Users/kai/.pyenv/versions/3.7.7/lib/python3.7/site-packages (from HEBO==0.3.2) (1.9.0)
Requirement already satisfied: numpy>=1.16 in /Users/kai/.pyenv/versions/3.7.7/lib/python3.7/site-packages (from HEBO==0.3.2) (1.21.6)
Collecting catboost>=0.24.4
  Downloading catboost-1.0.6-cp37-none-macosx_10_6_universal2.whl (21.9 MB)
     ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 21.9/21.9 MB 5.2 MB/s eta 0:00:0000:0100:01
Collecting pymoo>=0.5.0
  Downloading pymoo-0.5.0-cp37-cp37m-macosx_10_14_x86_64.whl (1.1 MB)
     ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 1.1/1.1 MB 3.9 MB/s eta 0:00:0000:0100:01
Requirement already satisfied: pandas>=1.0.1 in /Users/kai/.pyenv/versions/3.7.7/lib/python3.7/site-packages (from HEBO==0.3.2) (1.3.5)
Requirement already satisfied: matplotlib in /Users/kai/.pyenv/versions/3.7.7/lib/python3.7/site-packages (from catboost>=0.24.4->HEBO==0.3.2) (3.5.0)
Collecting graphviz
  Downloading graphviz-0.20-py3-none-any.whl (46 kB)
     ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 47.0/47.0 kB 1.4 MB/s eta 0:00:00
Requirement already satisfied: plotly in /Users/kai/.pyenv/versions/3.7.7/lib/python3.7/site-packages (from catboost>=0.24.4->HEBO==0.3.2) (5.6.0)
Requirement already satisfied: scipy in /Users/kai/.pyenv/versions/3.7.7/lib/python3.7/site-packages (from catboost>=0.24.4->HEBO==0.3.2) (1.4.1)
Requirement already satisfied: six in /Users/kai/.pyenv/versions/3.7.7/lib/python3.7/site-packages (from catboost>=0.24.4->HEBO==0.3.2) (1.16.0)
Requirement already satisfied: cython>=0.29 in /Users/kai/.pyenv/versions/3.7.7/lib/python3.7/site-packages (from GPy>=1.9.9->HEBO==0.3.2) (0.29.30)
Collecting paramz>=0.9.0
  Using cached paramz-0.9.5-py3-none-any.whl
Requirement already satisfied: python-dateutil>=2.7.3 in /Users/kai/.pyenv/versions/3.7.7/lib/python3.7/site-packages (from pandas>=1.0.1->HEBO==0.3.2) (2.8.2)
Requirement already satisfied: pytz>=2017.3 in /Users/kai/.pyenv/versions/3.7.7/lib/python3.7/site-packages (from pandas>=1.0.1->HEBO==0.3.2) (2022.1)
Collecting cma==2.7
  Using cached cma-2.7.0-py2.py3-none-any.whl (239 kB)
Collecting autograd>=1.3
  Downloading autograd-1.4-py3-none-any.whl (48 kB)
     ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 48.8/48.8 kB 1.5 MB/s eta 0:00:00
Requirement already satisfied: joblib>=0.11 in /Users/kai/.pyenv/versions/3.7.7/lib/python3.7/site-packages (from scikit-learn>=0.22->HEBO==0.3.2) (1.1.0)
Requirement already satisfied: threadpoolctl>=2.0.0 in /Users/kai/.pyenv/versions/3.7.7/lib/python3.7/site-packages (from scikit-learn>=0.22->HEBO==0.3.2) (3.0.0)
Requirement already satisfied: typing-extensions in /Users/kai/.pyenv/versions/3.7.7/lib/python3.7/site-packages (from torch>=1.9.0->HEBO==0.3.2) (4.1.1)
Requirement already satisfied: future>=0.15.2 in /Users/kai/.pyenv/versions/3.7.7/lib/python3.7/site-packages (from autograd>=1.3->pymoo>=0.5.0->HEBO==0.3.2) (0.18.2)
Requirement already satisfied: pillow>=6.2.0 in /Users/kai/.pyenv/versions/3.7.7/lib/python3.7/site-packages (from matplotlib->catboost>=0.24.4->HEBO==0.3.2) (9.1.0)
Requirement already satisfied: kiwisolver>=1.0.1 in /Users/kai/.pyenv/versions/3.7.7/lib/python3.7/site-packages (from matplotlib->catboost>=0.24.4->HEBO==0.3.2) (1.3.2)
Requirement already satisfied: packaging>=20.0 in /Users/kai/.pyenv/versions/3.7.7/lib/python3.7/site-packages (from matplotlib->catboost>=0.24.4->HEBO==0.3.2) (21.3)
Requirement already satisfied: setuptools-scm>=4 in /Users/kai/.pyenv/versions/3.7.7/lib/python3.7/site-packages (from matplotlib->catboost>=0.24.4->HEBO==0.3.2) (6.3.2)
Requirement already satisfied: pyparsing>=2.2.1 in /Users/kai/.pyenv/versions/3.7.7/lib/python3.7/site-packages (from matplotlib->catboost>=0.24.4->HEBO==0.3.2) (2.4.7)
Requirement already satisfied: cycler>=0.10 in /Users/kai/.pyenv/versions/3.7.7/lib/python3.7/site-packages (from matplotlib->catboost>=0.24.4->HEBO==0.3.2) (0.11.0)
Requirement already satisfied: fonttools>=4.22.0 in /Users/kai/.pyenv/versions/3.7.7/lib/python3.7/site-packages (from matplotlib->catboost>=0.24.4->HEBO==0.3.2) (4.28.2)
Requirement already satisfied: decorator>=4.0.10 in /Users/kai/.pyenv/versions/3.7.7/lib/python3.7/site-packages (from paramz>=0.9.0->GPy>=1.9.9->HEBO==0.3.2) (5.1.1)
Requirement already satisfied: tenacity>=6.2.0 in /Users/kai/.pyenv/versions/3.7.7/lib/python3.7/site-packages (from plotly->catboost>=0.24.4->HEBO==0.3.2) (8.0.1)
Requirement already satisfied: setuptools in /Users/kai/.pyenv/versions/3.7.7/lib/python3.7/site-packages (from setuptools-scm>=4->matplotlib->catboost>=0.24.4->HEBO==0.3.2) (59.5.0)
Requirement already satisfied: tomli>=1.0.0 in /Users/kai/.pyenv/versions/3.7.7/lib/python3.7/site-packages (from setuptools-scm>=4->matplotlib->catboost>=0.24.4->HEBO==0.3.2) (1.2.3)
Installing collected packages: cma, graphviz, autograd, paramz, GPy, pymoo, catboost, HEBO
  Attempting uninstall: cma
    Found existing installation: cma 3.2.2
    Uninstalling cma-3.2.2:
      Successfully uninstalled cma-3.2.2
Successfully installed GPy-1.10.0 HEBO-0.3.2 autograd-1.4 catboost-1.0.6 cma-2.7.0 graphviz-0.20 paramz-0.9.5 pymoo-0.5.0
WARNING: There was an error checking the latest version of pip.

Click below to see all the imports we need for this example. You can also launch directly into a Binder instance to run this notebook yourself. Just click on the rocket symbol at the top of the navigation.

In [2]:

import time

import ray
from ray import tune
from ray.air import session
from ray.tune.search.hebo import HEBOSearch

Let's start by defining a simple evaluation function. We artificially sleep for a bit (0.1 seconds) to simulate a long-running ML experiment. This setup assumes that we're running multiple steps of an experiment and try to tune two hyperparameters, namely width and height, and activation.

In [3]:

def evaluate(step, width, height, activation):
    time.sleep(0.1)
    activation_boost = 10 if activation=="relu" else 1
    return (0.1 + width * step / 100) ** (-1) + height * 0.1 + activation_boost

Next, our objective function takes a Tune config, evaluates the score of your experiment in a training loop, and uses session.report to report the score back to Tune.

In [4]:

def objective(config):
    for step in range(config["steps"]):
        score = evaluate(step, config["width"], config["height"], config["activation"])
        session.report({"iterations": step, "mean_loss": score})

In [5]:

ray.init(configure_logging=False)

Out[5]:

Ray

Python version:	3.7.7
Ray version:	3.0.0.dev0
Dashboard:	http://127.0.0.1:8266

While defining the search algorithm, we may choose to provide an initial set of hyperparameters that we believe are especially promising or informative, and pass this information as a helpful starting point for the HyperOptSearch object.

We also set the maximum concurrent trials to 8.

In [6]:

previously_run_params = [
    {"width": 10, "height": 0, "activation": "relu"},
    {"width": 15, "height": -20, "activation": "tanh"},
]

known_rewards = [-189, -1144]

max_concurrent = 8

algo = HEBOSearch(
    metric="mean_loss",
    mode="min",
    points_to_evaluate=previously_run_params,
    evaluated_rewards=known_rewards,
    random_state_seed=123,
    max_concurrent=max_concurrent,
)

The number of samples is the number of hyperparameter combinations that will be tried out. This Tune run is set to 1000 samples. (you can decrease this if it takes too long on your machine).

In [7]:

num_samples = 1000

In [8]:

# If 1000 samples take too long, you can reduce this number.
# We override this number here for our smoke tests.
num_samples = 10

Next we define a search space. The critical assumption is that the optimal hyperparamters live within this space. Yet, if the space is very large, then those hyperparameters may be difficult to find in a short amount of time.

In [9]:

search_config = {
    "steps": 100,
    "width": tune.uniform(0, 20),
    "height": tune.uniform(-100, 100),
    "activation": tune.choice(["relu", "tanh"])
}

Finally, we run the experiment to "min"imize the "mean_loss" of the objective by searching search_config via algo, num_samples times. This previous sentence is fully characterizes the search problem we aim to solve. With this in mind, notice how efficient it is to execute tuner.fit().

In [10]:

tuner = tune.Tuner(
    objective,
    tune_config=tune.TuneConfig(
        metric="mean_loss",
        mode="min",
        search_alg=algo,
        num_samples=num_samples,
    ),
    param_space=search_config,
)
results = tuner.fit()

== Status ==
Current time: 2022-07-22 15:35:11 (running for 00:00:36.78)
Memory usage on this node: 10.2/16.0 GiB
Using FIFO scheduling algorithm.
Resources requested: 0/16 CPUs, 0/0 GPUs, 0.0/5.3 GiB heap, 0.0/2.0 GiB objects
Current best trial: 72267d26 with mean_loss=-8.280721582416527 and parameters={'steps': 100, 'width': 16.267813332265522, 'height': -93.42430416543701, 'activation': 'tanh'}
Result logdir: /Users/kai/ray_results/objective_2022-07-22_15-34-34
Number of trials: 10/10 (10 TERMINATED)

Trial name	status	loc	activation	height	width	loss	iter	total time (s)	iterations	neg_mean_loss
objective_67ec1a0a	TERMINATED	127.0.0.1:47498	relu	-100	0	10	100	11.4386	99	-10
objective_69ac3226	TERMINATED	127.0.0.1:47512	relu	0	10	10.1	100	10.9018	99	-10.1
objective_69ada8d6	TERMINATED	127.0.0.1:47513	relu	-50	15	5.06689	100	10.7768	99	-5.06689
objective_69af2530	TERMINATED	127.0.0.1:47514	tanh	50	5	6.19802	100	10.9312	99	-6.19802
objective_69b0a8a6	TERMINATED	127.0.0.1:47515	tanh	-25	7.5	-1.36711	100	10.7948	99	1.36711
objective_69b2375c	TERMINATED	127.0.0.1:47516	relu	75	17.5	17.5574	100	10.8966	99	-17.5574
objective_69b3bb9a	TERMINATED	127.0.0.1:47517	tanh	-75	12.5	-6.41984	100	10.9022	99	6.41984
objective_69b58f60	TERMINATED	127.0.0.1:47519	relu	25	2.5	12.8883	100	10.8995	99	-12.8883
objective_72267d26	TERMINATED	127.0.0.1:47563	tanh	-93.4243	16.2678	-8.28072	100	10.7101	99	8.28072
objective_75ed3e0e	TERMINATED	127.0.0.1:47568	tanh	28.8058	15.0428	3.94728	100	10.7472	99	-3.94728

Result for objective_67ec1a0a:
  date: 2022-07-22_15-34-37
  done: false
  experiment_id: b2cc3485f1024cbbbb5947a9acd341e9
  hostname: Kais-MacBook-Pro.local
  iterations: 0
  iterations_since_restore: 1
  mean_loss: 10.0
  neg_mean_loss: -10.0
  node_ip: 127.0.0.1
  pid: 47498
  time_since_restore: 0.10423088073730469
  time_this_iter_s: 0.10423088073730469
  time_total_s: 0.10423088073730469
  timestamp: 1658500477
  timesteps_since_restore: 0
  training_iteration: 1
  trial_id: 67ec1a0a
  warmup_time: 0.0028820037841796875
  
Result for objective_69ada8d6:
  date: 2022-07-22_15-34-40
  done: false
  experiment_id: 9ca732d0f466455cbaa1da6f553a17ab
  hostname: Kais-MacBook-Pro.local
  iterations: 0
  iterations_since_restore: 1
  mean_loss: 15.0
  neg_mean_loss: -15.0
  node_ip: 127.0.0.1
  pid: 47513
  time_since_restore: 0.10410189628601074
  time_this_iter_s: 0.10410189628601074
  time_total_s: 0.10410189628601074
  timestamp: 1658500480
  timesteps_since_restore: 0
  training_iteration: 1
  trial_id: 69ada8d6
  warmup_time: 0.00498199462890625
  
Result for objective_69ac3226:
  date: 2022-07-22_15-34-40
  done: false
  experiment_id: 5fa0de7eaf624b22bf76f0407a5dc3cd
  hostname: Kais-MacBook-Pro.local
  iterations: 0
  iterations_since_restore: 1
  mean_loss: 20.0
  neg_mean_loss: -20.0
  node_ip: 127.0.0.1
  pid: 47512
  time_since_restore: 0.10359311103820801
  time_this_iter_s: 0.10359311103820801
  time_total_s: 0.10359311103820801
  timestamp: 1658500480
  timesteps_since_restore: 0
  training_iteration: 1
  trial_id: 69ac3226
  warmup_time: 0.007561922073364258
  
Result for objective_69af2530:
  date: 2022-07-22_15-34-40
  done: false
  experiment_id: 4c5fc14d64b04ec2b071fb751a9c6bde
  hostname: Kais-MacBook-Pro.local
  iterations: 0
  iterations_since_restore: 1
  mean_loss: 16.0
  neg_mean_loss: -16.0
  node_ip: 127.0.0.1
  pid: 47514
  time_since_restore: 0.1039130687713623
  time_this_iter_s: 0.1039130687713623
  time_total_s: 0.1039130687713623
  timestamp: 1658500480
  timesteps_since_restore: 0
  training_iteration: 1
  trial_id: 69af2530
  warmup_time: 0.002995014190673828
  
Result for objective_69b0a8a6:
  date: 2022-07-22_15-34-40
  done: false
  experiment_id: 013499ab54ed4a4f92666a27945d673e
  hostname: Kais-MacBook-Pro.local
  iterations: 0
  iterations_since_restore: 1
  mean_loss: 8.5
  neg_mean_loss: -8.5
  node_ip: 127.0.0.1
  pid: 47515
  time_since_restore: 0.10396409034729004
  time_this_iter_s: 0.10396409034729004
  time_total_s: 0.10396409034729004
  timestamp: 1658500480
  timesteps_since_restore: 0
  training_iteration: 1
  trial_id: 69b0a8a6
  warmup_time: 0.00310516357421875
  
Result for objective_69b2375c:
  date: 2022-07-22_15-34-40
  done: false
  experiment_id: 7e5aeb8fba3a42e6ae6471eedfc75fd2
  hostname: Kais-MacBook-Pro.local
  iterations: 0
  iterations_since_restore: 1
  mean_loss: 27.5
  neg_mean_loss: -27.5
  node_ip: 127.0.0.1
  pid: 47516
  time_since_restore: 0.10422396659851074
  time_this_iter_s: 0.10422396659851074
  time_total_s: 0.10422396659851074
  timestamp: 1658500480
  timesteps_since_restore: 0
  training_iteration: 1
  trial_id: 69b2375c
  warmup_time: 0.0030279159545898438
  
Result for objective_69b58f60:
  date: 2022-07-22_15-34-40
  done: false
  experiment_id: 02a11d6943f04adaaf8a7c50cbcee0dd
  hostname: Kais-MacBook-Pro.local
  iterations: 0
  iterations_since_restore: 1
  mean_loss: 22.5
  neg_mean_loss: -22.5
  node_ip: 127.0.0.1
  pid: 47519
  time_since_restore: 0.1043708324432373
  time_this_iter_s: 0.1043708324432373
  time_total_s: 0.1043708324432373
  timestamp: 1658500480
  timesteps_since_restore: 0
  training_iteration: 1
  trial_id: 69b58f60
  warmup_time: 0.0027899742126464844
  
Result for objective_69b3bb9a:
  date: 2022-07-22_15-34-40
  done: false
  experiment_id: a2d184f9ca934a768b78fbb438dbf28f
  hostname: Kais-MacBook-Pro.local
  iterations: 0
  iterations_since_restore: 1
  mean_loss: 3.5
  neg_mean_loss: -3.5
  node_ip: 127.0.0.1
  pid: 47517
  time_since_restore: 0.10400700569152832
  time_this_iter_s: 0.10400700569152832
  time_total_s: 0.10400700569152832
  timestamp: 1658500480
  timesteps_since_restore: 0
  training_iteration: 1
  trial_id: 69b3bb9a
  warmup_time: 0.002424955368041992
  
Result for objective_67ec1a0a:
  date: 2022-07-22_15-34-42
  done: false
  experiment_id: b2cc3485f1024cbbbb5947a9acd341e9
  hostname: Kais-MacBook-Pro.local
  iterations: 41
  iterations_since_restore: 42
  mean_loss: 10.0
  neg_mean_loss: -10.0
  node_ip: 127.0.0.1
  pid: 47498
  time_since_restore: 5.111451864242554
  time_this_iter_s: 0.10680818557739258
  time_total_s: 5.111451864242554
  timestamp: 1658500482
  timesteps_since_restore: 0
  training_iteration: 42
  trial_id: 67ec1a0a
  warmup_time: 0.0028820037841796875
  
Result for objective_69ac3226:
  date: 2022-07-22_15-34-45
  done: false
  experiment_id: 5fa0de7eaf624b22bf76f0407a5dc3cd
  hostname: Kais-MacBook-Pro.local
  iterations: 46
  iterations_since_restore: 47
  mean_loss: 10.212765957446809
  neg_mean_loss: -10.212765957446809
  node_ip: 127.0.0.1
  pid: 47512
  time_since_restore: 5.153754234313965
  time_this_iter_s: 0.10711407661437988
  time_total_s: 5.153754234313965
  timestamp: 1658500485
  timesteps_since_restore: 0
  training_iteration: 47
  trial_id: 69ac3226
  warmup_time: 0.007561922073364258
  
Result for objective_69af2530:
  date: 2022-07-22_15-34-45
  done: false
  experiment_id: 4c5fc14d64b04ec2b071fb751a9c6bde
  hostname: Kais-MacBook-Pro.local
  iterations: 46
  iterations_since_restore: 47
  mean_loss: 6.416666666666667
  neg_mean_loss: -6.416666666666667
  node_ip: 127.0.0.1
  pid: 47514
  time_since_restore: 5.158767938613892
  time_this_iter_s: 0.10651683807373047
  time_total_s: 5.158767938613892
  timestamp: 1658500485
  timesteps_since_restore: 0
  training_iteration: 47
  trial_id: 69af2530
  warmup_time: 0.002995014190673828
  
Result for objective_69b3bb9a:
  date: 2022-07-22_15-34-45
  done: false
  experiment_id: a2d184f9ca934a768b78fbb438dbf28f
  hostname: Kais-MacBook-Pro.local
  iterations: 46
  iterations_since_restore: 47
  mean_loss: -6.329059829059829
  neg_mean_loss: 6.329059829059829
  node_ip: 127.0.0.1
  pid: 47517
  time_since_restore: 5.1304240226745605
  time_this_iter_s: 0.10833311080932617
  time_total_s: 5.1304240226745605
  timestamp: 1658500485
  timesteps_since_restore: 0
  training_iteration: 47
  trial_id: 69b3bb9a
  warmup_time: 0.002424955368041992
  
Result for objective_69b58f60:
  date: 2022-07-22_15-34-45
  done: false
  experiment_id: 02a11d6943f04adaaf8a7c50cbcee0dd
  hostname: Kais-MacBook-Pro.local
  iterations: 46
  iterations_since_restore: 47
  mean_loss: 13.3
  neg_mean_loss: -13.3
  node_ip: 127.0.0.1
  pid: 47519
  time_since_restore: 5.138491868972778
  time_this_iter_s: 0.10869002342224121
  time_total_s: 5.138491868972778
  timestamp: 1658500485
  timesteps_since_restore: 0
  training_iteration: 47
  trial_id: 69b58f60
  warmup_time: 0.0027899742126464844
  
Result for objective_69b2375c:
  date: 2022-07-22_15-34-45
  done: false
  experiment_id: 7e5aeb8fba3a42e6ae6471eedfc75fd2
  hostname: Kais-MacBook-Pro.local
  iterations: 46
  iterations_since_restore: 47
  mean_loss: 17.62269938650307
  neg_mean_loss: -17.62269938650307
  node_ip: 127.0.0.1
  pid: 47516
  time_since_restore: 5.13613486289978
  time_this_iter_s: 0.10693097114562988
  time_total_s: 5.13613486289978
  timestamp: 1658500485
  timesteps_since_restore: 0
  training_iteration: 47
  trial_id: 69b2375c
  warmup_time: 0.0030279159545898438
  
Result for objective_69ada8d6:
  date: 2022-07-22_15-34-45
  done: false
  experiment_id: 9ca732d0f466455cbaa1da6f553a17ab
  hostname: Kais-MacBook-Pro.local
  iterations: 47
  iterations_since_restore: 48
  mean_loss: 5.13986013986014
  neg_mean_loss: -5.13986013986014
  node_ip: 127.0.0.1
  pid: 47513
  time_since_restore: 5.1575539112091064
  time_this_iter_s: 0.10637593269348145
  time_total_s: 5.1575539112091064
  timestamp: 1658500485
  timesteps_since_restore: 0
  training_iteration: 48
  trial_id: 69ada8d6
  warmup_time: 0.00498199462890625
  
Result for objective_69b0a8a6:
  date: 2022-07-22_15-34-45
  done: false
  experiment_id: 013499ab54ed4a4f92666a27945d673e
  hostname: Kais-MacBook-Pro.local
  iterations: 47
  iterations_since_restore: 48
  mean_loss: -1.2241379310344827
  neg_mean_loss: 1.2241379310344827
  node_ip: 127.0.0.1
  pid: 47515
  time_since_restore: 5.211113929748535
  time_this_iter_s: 0.10501360893249512
  time_total_s: 5.211113929748535
  timestamp: 1658500485
  timesteps_since_restore: 0
  training_iteration: 48
  trial_id: 69b0a8a6
  warmup_time: 0.00310516357421875
  
Result for objective_67ec1a0a:
  date: 2022-07-22_15-34-47
  done: false
  experiment_id: b2cc3485f1024cbbbb5947a9acd341e9
  hostname: Kais-MacBook-Pro.local
  iterations: 87
  iterations_since_restore: 88
  mean_loss: 10.0
  neg_mean_loss: -10.0
  node_ip: 127.0.0.1
  pid: 47498
  time_since_restore: 10.140707731246948
  time_this_iter_s: 0.10805296897888184
  time_total_s: 10.140707731246948
  timestamp: 1658500487
  timesteps_since_restore: 0
  training_iteration: 88
  trial_id: 67ec1a0a
  warmup_time: 0.0028820037841796875
  
Result for objective_67ec1a0a:
  date: 2022-07-22_15-34-48
  done: true
  experiment_id: b2cc3485f1024cbbbb5947a9acd341e9
  experiment_tag: 1_activation=relu,height=-100.0000,steps=100,width=0.0000
  hostname: Kais-MacBook-Pro.local
  iterations: 99
  iterations_since_restore: 100
  mean_loss: 10.0
  neg_mean_loss: -10.0
  node_ip: 127.0.0.1
  pid: 47498
  time_since_restore: 11.438636064529419
  time_this_iter_s: 0.1079721450805664
  time_total_s: 11.438636064529419
  timestamp: 1658500488
  timesteps_since_restore: 0
  training_iteration: 100
  trial_id: 67ec1a0a
  warmup_time: 0.0028820037841796875
  
Result for objective_69b3bb9a:
  date: 2022-07-22_15-34-50
  done: false
  experiment_id: a2d184f9ca934a768b78fbb438dbf28f
  hostname: Kais-MacBook-Pro.local
  iterations: 92
  iterations_since_restore: 93
  mean_loss: -6.413793103448276
  neg_mean_loss: 6.413793103448276
  node_ip: 127.0.0.1
  pid: 47517
  time_since_restore: 10.136809825897217
  time_this_iter_s: 0.10945367813110352
  time_total_s: 10.136809825897217
  timestamp: 1658500490
  timesteps_since_restore: 0
  training_iteration: 93
  trial_id: 69b3bb9a
  warmup_time: 0.002424955368041992
  
Result for objective_69ac3226:
  date: 2022-07-22_15-34-50
  done: false
  experiment_id: 5fa0de7eaf624b22bf76f0407a5dc3cd
  hostname: Kais-MacBook-Pro.local
  iterations: 93
  iterations_since_restore: 94
  mean_loss: 10.106382978723405
  neg_mean_loss: -10.106382978723405
  node_ip: 127.0.0.1
  pid: 47512
  time_since_restore: 10.222928285598755
  time_this_iter_s: 0.10617327690124512
  time_total_s: 10.222928285598755
  timestamp: 1658500490
  timesteps_since_restore: 0
  training_iteration: 94
  trial_id: 69ac3226
  warmup_time: 0.007561922073364258
  
Result for objective_69ada8d6:
  date: 2022-07-22_15-34-50
  done: false
  experiment_id: 9ca732d0f466455cbaa1da6f553a17ab
  hostname: Kais-MacBook-Pro.local
  iterations: 94
  iterations_since_restore: 95
  mean_loss: 5.070422535211268
  neg_mean_loss: -5.070422535211268
  node_ip: 127.0.0.1
  pid: 47513
  time_since_restore: 10.23993706703186
  time_this_iter_s: 0.10660290718078613
  time_total_s: 10.23993706703186
  timestamp: 1658500490
  timesteps_since_restore: 0
  training_iteration: 95
  trial_id: 69ada8d6
  warmup_time: 0.00498199462890625
  
Result for objective_69b2375c:
  date: 2022-07-22_15-34-50
  done: false
  experiment_id: 7e5aeb8fba3a42e6ae6471eedfc75fd2
  hostname: Kais-MacBook-Pro.local
  iterations: 93
  iterations_since_restore: 94
  mean_loss: 17.561068702290076
  neg_mean_loss: -17.561068702290076
  node_ip: 127.0.0.1
  pid: 47516
  time_since_restore: 10.21057415008545
  time_this_iter_s: 0.10399723052978516
  time_total_s: 10.21057415008545
  timestamp: 1658500490
  timesteps_since_restore: 0
  training_iteration: 94
  trial_id: 69b2375c
  warmup_time: 0.0030279159545898438
  
Result for objective_69b58f60:
  date: 2022-07-22_15-34-50
  done: false
  experiment_id: 02a11d6943f04adaaf8a7c50cbcee0dd
  hostname: Kais-MacBook-Pro.local
  iterations: 93
  iterations_since_restore: 94
  mean_loss: 12.912371134020619
  neg_mean_loss: -12.912371134020619
  node_ip: 127.0.0.1
  pid: 47519
  time_since_restore: 10.214950799942017
  time_this_iter_s: 0.10687804222106934
  time_total_s: 10.214950799942017
  timestamp: 1658500490
  timesteps_since_restore: 0
  training_iteration: 94
  trial_id: 69b58f60
  warmup_time: 0.0027899742126464844
  
Result for objective_69af2530:
  date: 2022-07-22_15-34-50
  done: false
  experiment_id: 4c5fc14d64b04ec2b071fb751a9c6bde
  hostname: Kais-MacBook-Pro.local
  iterations: 93
  iterations_since_restore: 94
  mean_loss: 6.2105263157894735
  neg_mean_loss: -6.2105263157894735
  node_ip: 127.0.0.1
  pid: 47514
  time_since_restore: 10.267423152923584
  time_this_iter_s: 0.10761213302612305
  time_total_s: 10.267423152923584
  timestamp: 1658500490
  timesteps_since_restore: 0
  training_iteration: 94
  trial_id: 69af2530
  warmup_time: 0.002995014190673828
  
Result for objective_69b0a8a6:
  date: 2022-07-22_15-34-50
  done: false
  experiment_id: 013499ab54ed4a4f92666a27945d673e
  hostname: Kais-MacBook-Pro.local
  iterations: 94
  iterations_since_restore: 95
  mean_loss: -1.36013986013986
  neg_mean_loss: 1.36013986013986
  node_ip: 127.0.0.1
  pid: 47515
  time_since_restore: 10.256262063980103
  time_this_iter_s: 0.10606575012207031
  time_total_s: 10.256262063980103
  timestamp: 1658500490
  timesteps_since_restore: 0
  training_iteration: 95
  trial_id: 69b0a8a6
  warmup_time: 0.00310516357421875
  
Result for objective_69ada8d6:
  date: 2022-07-22_15-34-51
  done: true
  experiment_id: 9ca732d0f466455cbaa1da6f553a17ab
  experiment_tag: 3_activation=relu,height=-50.0000,steps=100,width=15.0000
  hostname: Kais-MacBook-Pro.local
  iterations: 99
  iterations_since_restore: 100
  mean_loss: 5.066889632107023
  neg_mean_loss: -5.066889632107023
  node_ip: 127.0.0.1
  pid: 47513
  time_since_restore: 10.77684497833252
  time_this_iter_s: 0.10641121864318848
  time_total_s: 10.77684497833252
  timestamp: 1658500491
  timesteps_since_restore: 0
  training_iteration: 100
  trial_id: 69ada8d6
  warmup_time: 0.00498199462890625
  
Result for objective_69b0a8a6:
  date: 2022-07-22_15-34-51
  done: true
  experiment_id: 013499ab54ed4a4f92666a27945d673e
  experiment_tag: 5_activation=tanh,height=-25.0000,steps=100,width=7.5000
  hostname: Kais-MacBook-Pro.local
  iterations: 99
  iterations_since_restore: 100
  mean_loss: -1.367109634551495
  neg_mean_loss: 1.367109634551495
  node_ip: 127.0.0.1
  pid: 47515
  time_since_restore: 10.794761180877686
  time_this_iter_s: 0.10670304298400879
  time_total_s: 10.794761180877686
  timestamp: 1658500491
  timesteps_since_restore: 0
  training_iteration: 100
  trial_id: 69b0a8a6
  warmup_time: 0.00310516357421875
  
Result for objective_69ac3226:
  date: 2022-07-22_15-34-51
  done: true
  experiment_id: 5fa0de7eaf624b22bf76f0407a5dc3cd
  experiment_tag: 2_activation=relu,height=0.0000,steps=100,width=10.0000
  hostname: Kais-MacBook-Pro.local
  iterations: 99
  iterations_since_restore: 100
  mean_loss: 10.1
  neg_mean_loss: -10.1
  node_ip: 127.0.0.1
  pid: 47512
  time_since_restore: 10.901827096939087
  time_this_iter_s: 0.13848495483398438
  time_total_s: 10.901827096939087
  timestamp: 1658500491
  timesteps_since_restore: 0
  training_iteration: 100
  trial_id: 69ac3226
  warmup_time: 0.007561922073364258
  
Result for objective_69b2375c:
  date: 2022-07-22_15-34-51
  done: true
  experiment_id: 7e5aeb8fba3a42e6ae6471eedfc75fd2
  experiment_tag: 6_activation=relu,height=75.0000,steps=100,width=17.5000
  hostname: Kais-MacBook-Pro.local
  iterations: 99
  iterations_since_restore: 100
  mean_loss: 17.55738880918221
  neg_mean_loss: -17.55738880918221
  node_ip: 127.0.0.1
  pid: 47516
  time_since_restore: 10.896636962890625
  time_this_iter_s: 0.14625000953674316
  time_total_s: 10.896636962890625
  timestamp: 1658500491
  timesteps_since_restore: 0
  training_iteration: 100
  trial_id: 69b2375c
  warmup_time: 0.0030279159545898438
  
Result for objective_69b3bb9a:
  date: 2022-07-22_15-34-51
  done: true
  experiment_id: a2d184f9ca934a768b78fbb438dbf28f
  experiment_tag: 7_activation=tanh,height=-75.0000,steps=100,width=12.5000
  hostname: Kais-MacBook-Pro.local
  iterations: 99
  iterations_since_restore: 100
  mean_loss: -6.419839679358717
  neg_mean_loss: 6.419839679358717
  node_ip: 127.0.0.1
  pid: 47517
  time_since_restore: 10.902234077453613
  time_this_iter_s: 0.12049722671508789
  time_total_s: 10.902234077453613
  timestamp: 1658500491
  timesteps_since_restore: 0
  training_iteration: 100
  trial_id: 69b3bb9a
  warmup_time: 0.002424955368041992
  
Result for objective_69b58f60:
  date: 2022-07-22_15-34-51
  done: true
  experiment_id: 02a11d6943f04adaaf8a7c50cbcee0dd
  experiment_tag: 8_activation=relu,height=25.0000,steps=100,width=2.5000
  hostname: Kais-MacBook-Pro.local
  iterations: 99
  iterations_since_restore: 100
  mean_loss: 12.888349514563107
  neg_mean_loss: -12.888349514563107
  node_ip: 127.0.0.1
  pid: 47519
  time_since_restore: 10.899547815322876
  time_this_iter_s: 0.1467878818511963
  time_total_s: 10.899547815322876
  timestamp: 1658500491
  timesteps_since_restore: 0
  training_iteration: 100
  trial_id: 69b58f60
  warmup_time: 0.0027899742126464844
  
Result for objective_69af2530:
  date: 2022-07-22_15-34-51
  done: true
  experiment_id: 4c5fc14d64b04ec2b071fb751a9c6bde
  experiment_tag: 4_activation=tanh,height=50.0000,steps=100,width=5.0000
  hostname: Kais-MacBook-Pro.local
  iterations: 99
  iterations_since_restore: 100
  mean_loss: 6.198019801980198
  neg_mean_loss: -6.198019801980198
  node_ip: 127.0.0.1
  pid: 47514
  time_since_restore: 10.931232929229736
  time_this_iter_s: 0.12574982643127441
  time_total_s: 10.931232929229736
  timestamp: 1658500491
  timesteps_since_restore: 0
  training_iteration: 100
  trial_id: 69af2530
  warmup_time: 0.002995014190673828
  
Result for objective_72267d26:
  date: 2022-07-22_15-34-58
  done: false
  experiment_id: 05d0fd74bba34c209c3fb167e5aabb6e
  hostname: Kais-MacBook-Pro.local
  iterations: 0
  iterations_since_restore: 1
  mean_loss: 1.657569583456299
  neg_mean_loss: -1.657569583456299
  node_ip: 127.0.0.1
  pid: 47563
  time_since_restore: 0.10434603691101074
  time_this_iter_s: 0.10434603691101074
  time_total_s: 0.10434603691101074
  timestamp: 1658500498
  timesteps_since_restore: 0
  training_iteration: 1
  trial_id: 72267d26
  warmup_time: 0.0029430389404296875
  
Result for objective_75ed3e0e:
  date: 2022-07-22_15-35-00
  done: false
  experiment_id: 57d23b5e98454d9eb68f3dee5b5f2642
  hostname: Kais-MacBook-Pro.local
  iterations: 0
  iterations_since_restore: 1
  mean_loss: 13.88058437447561
  neg_mean_loss: -13.88058437447561
  node_ip: 127.0.0.1
  pid: 47568
  time_since_restore: 0.10134601593017578
  time_this_iter_s: 0.10134601593017578
  time_total_s: 0.10134601593017578
  timestamp: 1658500500
  timesteps_since_restore: 0
  training_iteration: 1
  trial_id: 75ed3e0e
  warmup_time: 0.002665996551513672
  
Result for objective_72267d26:
  date: 2022-07-22_15-35-03
  done: false
  experiment_id: 05d0fd74bba34c209c3fb167e5aabb6e
  hostname: Kais-MacBook-Pro.local
  iterations: 47
  iterations_since_restore: 48
  mean_loss: -8.213329397880102
  neg_mean_loss: 8.213329397880102
  node_ip: 127.0.0.1
  pid: 47563
  time_since_restore: 5.1267828941345215
  time_this_iter_s: 0.10927891731262207
  time_total_s: 5.1267828941345215
  timestamp: 1658500503
  timesteps_since_restore: 0
  training_iteration: 48
  trial_id: 72267d26
  warmup_time: 0.0029430389404296875
  
Result for objective_75ed3e0e:
  date: 2022-07-22_15-35-05
  done: false
  experiment_id: 57d23b5e98454d9eb68f3dee5b5f2642
  hostname: Kais-MacBook-Pro.local
  iterations: 47
  iterations_since_restore: 48
  mean_loss: 4.020052046405574
  neg_mean_loss: -4.020052046405574
  node_ip: 127.0.0.1
  pid: 47568
  time_since_restore: 5.158367156982422
  time_this_iter_s: 0.10702204704284668
  time_total_s: 5.158367156982422
  timestamp: 1658500505
  timesteps_since_restore: 0
  training_iteration: 48
  trial_id: 75ed3e0e
  warmup_time: 0.002665996551513672
  
Result for objective_72267d26:
  date: 2022-07-22_15-35-08
  done: false
  experiment_id: 05d0fd74bba34c209c3fb167e5aabb6e
  hostname: Kais-MacBook-Pro.local
  iterations: 94
  iterations_since_restore: 95
  mean_loss: -8.277460523241512
  neg_mean_loss: 8.277460523241512
  node_ip: 127.0.0.1
  pid: 47563
  time_since_restore: 10.168545961380005
  time_this_iter_s: 0.10625672340393066
  time_total_s: 10.168545961380005
  timestamp: 1658500508
  timesteps_since_restore: 0
  training_iteration: 95
  trial_id: 72267d26
  warmup_time: 0.0029430389404296875
  
Result for objective_72267d26:
  date: 2022-07-22_15-35-08
  done: true
  experiment_id: 05d0fd74bba34c209c3fb167e5aabb6e
  experiment_tag: 9_activation=tanh,height=-93.4243,steps=100,width=16.2678
  hostname: Kais-MacBook-Pro.local
  iterations: 99
  iterations_since_restore: 100
  mean_loss: -8.280721582416527
  neg_mean_loss: 8.280721582416527
  node_ip: 127.0.0.1
  pid: 47563
  time_since_restore: 10.71009612083435
  time_this_iter_s: 0.10849308967590332
  time_total_s: 10.71009612083435
  timestamp: 1658500508
  timesteps_since_restore: 0
  training_iteration: 100
  trial_id: 72267d26
  warmup_time: 0.0029430389404296875
  
Result for objective_75ed3e0e:
  date: 2022-07-22_15-35-10
  done: false
  experiment_id: 57d23b5e98454d9eb68f3dee5b5f2642
  hostname: Kais-MacBook-Pro.local
  iterations: 94
  iterations_since_restore: 95
  mean_loss: 3.950807906063858
  neg_mean_loss: -3.950807906063858
  node_ip: 127.0.0.1
  pid: 47568
  time_since_restore: 10.20597505569458
  time_this_iter_s: 0.10656380653381348
  time_total_s: 10.20597505569458
  timestamp: 1658500510
  timesteps_since_restore: 0
  training_iteration: 95
  trial_id: 75ed3e0e
  warmup_time: 0.002665996551513672
  
Result for objective_75ed3e0e:
  date: 2022-07-22_15-35-11
  done: true
  experiment_id: 57d23b5e98454d9eb68f3dee5b5f2642
  experiment_tag: 10_activation=tanh,height=28.8058,steps=100,width=15.0428
  hostname: Kais-MacBook-Pro.local
  iterations: 99
  iterations_since_restore: 100
  mean_loss: 3.947284919356474
  neg_mean_loss: -3.947284919356474
  node_ip: 127.0.0.1
  pid: 47568
  time_since_restore: 10.74724817276001
  time_this_iter_s: 0.10794186592102051
  time_total_s: 10.74724817276001
  timestamp: 1658500511
  timesteps_since_restore: 0
  training_iteration: 100
  trial_id: 75ed3e0e
  warmup_time: 0.002665996551513672

Here are the hyperparamters found to minimize the mean loss of the defined objective.

In [11]:

print("Best hyperparameters found were: ", results.get_best_result().config)

Best hyperparameters found were:  {'steps': 100, 'width': 16.267813332265522, 'height': -93.42430416543701, 'activation': 'tanh'}

In [12]:

ray.shutdown()