Notebook

Running Tune experiments with ZOOpt¶

In this tutorial we introduce ZOOpt, 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 ZOOpt optimization process - without sacrificing performance.

Zeroth-order optimization (ZOOpt) 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. Therefore, zeroth-order optimization is commonly referred to as "derivative-free optimization" and "black-box optimization". In this example we minimize a simple objective to briefly demonstrate the usage of ZOOpt with Ray Tune via ZOOptSearch. 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 ZOOpt website.

In [1]:

# !pip install ray[tune]
!pip install zoopt==0.4.1

Requirement already satisfied: zoopt==0.4.1 in /Users/kai/.pyenv/versions/3.7.7/lib/python3.7/site-packages (0.4.1)
Requirement already satisfied: matplotlib in /Users/kai/.pyenv/versions/3.7.7/lib/python3.7/site-packages (from zoopt==0.4.1) (3.5.0)
Requirement already satisfied: numpy in /Users/kai/.pyenv/versions/3.7.7/lib/python3.7/site-packages (from zoopt==0.4.1) (1.21.6)
Requirement already satisfied: cycler>=0.10 in /Users/kai/.pyenv/versions/3.7.7/lib/python3.7/site-packages (from matplotlib->zoopt==0.4.1) (0.11.0)
Requirement already satisfied: python-dateutil>=2.7 in /Users/kai/.pyenv/versions/3.7.7/lib/python3.7/site-packages (from matplotlib->zoopt==0.4.1) (2.8.2)
Requirement already satisfied: fonttools>=4.22.0 in /Users/kai/.pyenv/versions/3.7.7/lib/python3.7/site-packages (from matplotlib->zoopt==0.4.1) (4.28.2)
Requirement already satisfied: packaging>=20.0 in /Users/kai/.pyenv/versions/3.7.7/lib/python3.7/site-packages (from matplotlib->zoopt==0.4.1) (21.3)
Requirement already satisfied: pillow>=6.2.0 in /Users/kai/.pyenv/versions/3.7.7/lib/python3.7/site-packages (from matplotlib->zoopt==0.4.1) (9.1.0)
Requirement already satisfied: setuptools-scm>=4 in /Users/kai/.pyenv/versions/3.7.7/lib/python3.7/site-packages (from matplotlib->zoopt==0.4.1) (6.3.2)
Requirement already satisfied: kiwisolver>=1.0.1 in /Users/kai/.pyenv/versions/3.7.7/lib/python3.7/site-packages (from matplotlib->zoopt==0.4.1) (1.3.2)
Requirement already satisfied: pyparsing>=2.2.1 in /Users/kai/.pyenv/versions/3.7.7/lib/python3.7/site-packages (from matplotlib->zoopt==0.4.1) (2.4.7)
Requirement already satisfied: six>=1.5 in /Users/kai/.pyenv/versions/3.7.7/lib/python3.7/site-packages (from python-dateutil>=2.7->matplotlib->zoopt==0.4.1) (1.16.0)
Requirement already satisfied: setuptools in /Users/kai/.pyenv/versions/3.7.7/lib/python3.7/site-packages (from setuptools-scm>=4->matplotlib->zoopt==0.4.1) (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->zoopt==0.4.1) (1.2.3)
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.zoopt import ZOOptSearch
from zoopt import ValueType

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):
    time.sleep(0.1)
    return (0.1 + width * step / 100) ** (-1) + height * 0.1

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"])
        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

Next we define a search space. The critical assumption is that the optimal hyperparameters 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 [6]:

search_config = {
    "steps": 100,
    "width": tune.randint(0, 10),
    "height": tune.quniform(-10, 10, 1e-2),
    "activation": tune.choice(["relu, tanh"])
}

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 the search algorithm built from ZOOptSearch, constrained to a maximum of 8 concurrent trials via ZOOpt's internal "parallel_num".

In [9]:

zoopt_config = {
    "parallel_num": 8
}
algo = ZOOptSearch(
    algo="Asracos",  # only supports ASRacos currently
    budget=num_samples,
    **zoopt_config,
)

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()

Function checkpointing is disabled. This may result in unexpected behavior when using checkpointing features or certain schedulers. To enable, set the train function arguments to be `func(config, checkpoint_dir=None)`.

== Status ==
Current time: 2022-07-22 15:36:05 (running for 00:00:29.94)
Memory usage on this node: 11.1/16.0 GiB
Using FIFO scheduling algorithm.
Resources requested: 0/16 CPUs, 0/0 GPUs, 0.0/5.39 GiB heap, 0.0/2.0 GiB objects
Current best trial: 963faf2a with mean_loss=-0.39843708609271516 and parameters={'steps': 100, 'width': 6, 'height': -5.64, 'activation': 'relu, tanh'}
Result logdir: /Users/kai/ray_results/objective_2022-07-22_15-35-34
Number of trials: 7/10 (7 TERMINATED)

Trial name	status	loc	activation	height	width	loss	iter	total time (s)	iterations	neg_mean_loss
objective_8c72f588	TERMINATED	127.0.0.1:47662	relu, tanh	-3.94	0	9.606	100	10.9102	99	-9.606
objective_8e2f11ae	TERMINATED	127.0.0.1:47667	relu, tanh	-0.68	6	0.0975629	100	10.7479	99	-0.0975629
objective_8e30a596	TERMINATED	127.0.0.1:47668	relu, tanh	-5.84	0	9.416	100	10.7724	99	-9.416
objective_8e32324e	TERMINATED	127.0.0.1:47669	relu, tanh	-2.15	3	0.110733	100	10.7684	99	-0.110733
objective_963faf2a	TERMINATED	127.0.0.1:47689	relu, tanh	-5.64	6	-0.398437	100	10.8267	99	0.398437
objective_96417df0	TERMINATED	127.0.0.1:47690	relu, tanh	2.84	6	0.449563	100	10.821	99	-0.449563
objective_96435da0	TERMINATED	127.0.0.1:47691	relu, tanh	2.6	6	0.425563	100	10.7694	99	-0.425563

Result for objective_8c72f588:
  date: 2022-07-22_15-35-38
  done: false
  experiment_id: 3eb9bcef55e341b0970abd6c1f97eda7
  hostname: Kais-MacBook-Pro.local
  iterations: 0
  iterations_since_restore: 1
  mean_loss: 9.606
  neg_mean_loss: -9.606
  node_ip: 127.0.0.1
  pid: 47662
  time_since_restore: 0.10410094261169434
  time_this_iter_s: 0.10410094261169434
  time_total_s: 0.10410094261169434
  timestamp: 1658500538
  timesteps_since_restore: 0
  training_iteration: 1
  trial_id: 8c72f588
  warmup_time: 0.003092050552368164
  
Result for objective_8e30a596:
  date: 2022-07-22_15-35-41
  done: false
  experiment_id: d58453075b71453ab615e10ae9713072
  hostname: Kais-MacBook-Pro.local
  iterations: 0
  iterations_since_restore: 1
  mean_loss: 9.416
  neg_mean_loss: -9.416
  node_ip: 127.0.0.1
  pid: 47668
  time_since_restore: 0.1051950454711914
  time_this_iter_s: 0.1051950454711914
  time_total_s: 0.1051950454711914
  timestamp: 1658500541
  timesteps_since_restore: 0
  training_iteration: 1
  trial_id: 8e30a596
  warmup_time: 0.004169940948486328
  
Result for objective_8e32324e:
  date: 2022-07-22_15-35-41
  done: false
  experiment_id: 22c7ba8baa2644479b661e6d91e5fae8
  hostname: Kais-MacBook-Pro.local
  iterations: 0
  iterations_since_restore: 1
  mean_loss: 9.785
  neg_mean_loss: -9.785
  node_ip: 127.0.0.1
  pid: 47669
  time_since_restore: 0.10500001907348633
  time_this_iter_s: 0.10500001907348633
  time_total_s: 0.10500001907348633
  timestamp: 1658500541
  timesteps_since_restore: 0
  training_iteration: 1
  trial_id: 8e32324e
  warmup_time: 0.004729032516479492
  
Result for objective_8e2f11ae:
  date: 2022-07-22_15-35-41
  done: false
  experiment_id: c4d325574058491c8af3a0c869f0ebe5
  hostname: Kais-MacBook-Pro.local
  iterations: 0
  iterations_since_restore: 1
  mean_loss: 9.932
  neg_mean_loss: -9.932
  node_ip: 127.0.0.1
  pid: 47667
  time_since_restore: 0.10310196876525879
  time_this_iter_s: 0.10310196876525879
  time_total_s: 0.10310196876525879
  timestamp: 1658500541
  timesteps_since_restore: 0
  training_iteration: 1
  trial_id: 8e2f11ae
  warmup_time: 0.0029730796813964844
  
Result for objective_8c72f588:
  date: 2022-07-22_15-35-43
  done: false
  experiment_id: 3eb9bcef55e341b0970abd6c1f97eda7
  hostname: Kais-MacBook-Pro.local
  iterations: 45
  iterations_since_restore: 46
  mean_loss: 9.606
  neg_mean_loss: -9.606
  node_ip: 127.0.0.1
  pid: 47662
  time_since_restore: 5.112913131713867
  time_this_iter_s: 0.10695910453796387
  time_total_s: 5.112913131713867
  timestamp: 1658500543
  timesteps_since_restore: 0
  training_iteration: 46
  trial_id: 8c72f588
  warmup_time: 0.003092050552368164
  
Result for objective_8e30a596:
  date: 2022-07-22_15-35-46
  done: false
  experiment_id: d58453075b71453ab615e10ae9713072
  hostname: Kais-MacBook-Pro.local
  iterations: 47
  iterations_since_restore: 48
  mean_loss: 9.416
  neg_mean_loss: -9.416
  node_ip: 127.0.0.1
  pid: 47668
  time_since_restore: 5.1615166664123535
  time_this_iter_s: 0.10595178604125977
  time_total_s: 5.1615166664123535
  timestamp: 1658500546
  timesteps_since_restore: 0
  training_iteration: 48
  trial_id: 8e30a596
  warmup_time: 0.004169940948486328
  
Result for objective_8e2f11ae:
  date: 2022-07-22_15-35-46
  done: false
  experiment_id: c4d325574058491c8af3a0c869f0ebe5
  hostname: Kais-MacBook-Pro.local
  iterations: 47
  iterations_since_restore: 48
  mean_loss: 0.2744657534246575
  neg_mean_loss: -0.2744657534246575
  node_ip: 127.0.0.1
  pid: 47667
  time_since_restore: 5.1498119831085205
  time_this_iter_s: 0.10741090774536133
  time_total_s: 5.1498119831085205
  timestamp: 1658500546
  timesteps_since_restore: 0
  training_iteration: 48
  trial_id: 8e2f11ae
  warmup_time: 0.0029730796813964844
  
Result for objective_8e32324e:
  date: 2022-07-22_15-35-46
  done: false
  experiment_id: 22c7ba8baa2644479b661e6d91e5fae8
  hostname: Kais-MacBook-Pro.local
  iterations: 47
  iterations_since_restore: 48
  mean_loss: 0.44725165562913916
  neg_mean_loss: -0.44725165562913916
  node_ip: 127.0.0.1
  pid: 47669
  time_since_restore: 5.166383981704712
  time_this_iter_s: 0.1064291000366211
  time_total_s: 5.166383981704712
  timestamp: 1658500546
  timesteps_since_restore: 0
  training_iteration: 48
  trial_id: 8e32324e
  warmup_time: 0.004729032516479492
  
Result for objective_8c72f588:
  date: 2022-07-22_15-35-48
  done: false
  experiment_id: 3eb9bcef55e341b0970abd6c1f97eda7
  hostname: Kais-MacBook-Pro.local
  iterations: 92
  iterations_since_restore: 93
  mean_loss: 9.606
  neg_mean_loss: -9.606
  node_ip: 127.0.0.1
  pid: 47662
  time_since_restore: 10.156940937042236
  time_this_iter_s: 0.10845208168029785
  time_total_s: 10.156940937042236
  timestamp: 1658500548
  timesteps_since_restore: 0
  training_iteration: 93
  trial_id: 8c72f588
  warmup_time: 0.003092050552368164
  
Result for objective_8c72f588:
  date: 2022-07-22_15-35-49
  done: true
  experiment_id: 3eb9bcef55e341b0970abd6c1f97eda7
  experiment_tag: 1_activation=relu_tanh,height=-3.9400,steps=100,width=0
  hostname: Kais-MacBook-Pro.local
  iterations: 99
  iterations_since_restore: 100
  mean_loss: 9.606
  neg_mean_loss: -9.606
  node_ip: 127.0.0.1
  pid: 47662
  time_since_restore: 10.910246133804321
  time_this_iter_s: 0.1059122085571289
  time_total_s: 10.910246133804321
  timestamp: 1658500549
  timesteps_since_restore: 0
  training_iteration: 100
  trial_id: 8c72f588
  warmup_time: 0.003092050552368164
  
Result for objective_8e2f11ae:
  date: 2022-07-22_15-35-51
  done: false
  experiment_id: c4d325574058491c8af3a0c869f0ebe5
  hostname: Kais-MacBook-Pro.local
  iterations: 94
  iterations_since_restore: 95
  mean_loss: 0.10621602787456447
  neg_mean_loss: -0.10621602787456447
  node_ip: 127.0.0.1
  pid: 47667
  time_since_restore: 10.211436986923218
  time_this_iter_s: 0.10804891586303711
  time_total_s: 10.211436986923218
  timestamp: 1658500551
  timesteps_since_restore: 0
  training_iteration: 95
  trial_id: 8e2f11ae
  warmup_time: 0.0029730796813964844
  
Result for objective_8e32324e:
  date: 2022-07-22_15-35-51
  done: false
  experiment_id: 22c7ba8baa2644479b661e6d91e5fae8
  hostname: Kais-MacBook-Pro.local
  iterations: 94
  iterations_since_restore: 95
  mean_loss: 0.12746575342465752
  neg_mean_loss: -0.12746575342465752
  node_ip: 127.0.0.1
  pid: 47669
  time_since_restore: 10.228847980499268
  time_this_iter_s: 0.10761308670043945
  time_total_s: 10.228847980499268
  timestamp: 1658500551
  timesteps_since_restore: 0
  training_iteration: 95
  trial_id: 8e32324e
  warmup_time: 0.004729032516479492
  
Result for objective_8e30a596:
  date: 2022-07-22_15-35-51
  done: false
  experiment_id: d58453075b71453ab615e10ae9713072
  hostname: Kais-MacBook-Pro.local
  iterations: 94
  iterations_since_restore: 95
  mean_loss: 9.416
  neg_mean_loss: -9.416
  node_ip: 127.0.0.1
  pid: 47668
  time_since_restore: 10.231056928634644
  time_this_iter_s: 0.10677409172058105
  time_total_s: 10.231056928634644
  timestamp: 1658500551
  timesteps_since_restore: 0
  training_iteration: 95
  trial_id: 8e30a596
  warmup_time: 0.004169940948486328
  
Result for objective_8e2f11ae:
  date: 2022-07-22_15-35-52
  done: true
  experiment_id: c4d325574058491c8af3a0c869f0ebe5
  experiment_tag: 2_activation=relu_tanh,height=-0.6800,steps=100,width=6
  hostname: Kais-MacBook-Pro.local
  iterations: 99
  iterations_since_restore: 100
  mean_loss: 0.09756291390728478
  neg_mean_loss: -0.09756291390728478
  node_ip: 127.0.0.1
  pid: 47667
  time_since_restore: 10.747868061065674
  time_this_iter_s: 0.10819792747497559
  time_total_s: 10.747868061065674
  timestamp: 1658500552
  timesteps_since_restore: 0
  training_iteration: 100
  trial_id: 8e2f11ae
  warmup_time: 0.0029730796813964844
  
Result for objective_8e32324e:
  date: 2022-07-22_15-35-52
  done: true
  experiment_id: 22c7ba8baa2644479b661e6d91e5fae8
  experiment_tag: 4_activation=relu_tanh,height=-2.1500,steps=100,width=3
  hostname: Kais-MacBook-Pro.local
  iterations: 99
  iterations_since_restore: 100
  mean_loss: 0.11073289902280128
  neg_mean_loss: -0.11073289902280128
  node_ip: 127.0.0.1
  pid: 47669
  time_since_restore: 10.768368005752563
  time_this_iter_s: 0.10648989677429199
  time_total_s: 10.768368005752563
  timestamp: 1658500552
  timesteps_since_restore: 0
  training_iteration: 100
  trial_id: 8e32324e
  warmup_time: 0.004729032516479492
  
Result for objective_8e30a596:
  date: 2022-07-22_15-35-52
  done: true
  experiment_id: d58453075b71453ab615e10ae9713072
  experiment_tag: 3_activation=relu_tanh,height=-5.8400,steps=100,width=0
  hostname: Kais-MacBook-Pro.local
  iterations: 99
  iterations_since_restore: 100
  mean_loss: 9.416
  neg_mean_loss: -9.416
  node_ip: 127.0.0.1
  pid: 47668
  time_since_restore: 10.7723867893219
  time_this_iter_s: 0.10686278343200684
  time_total_s: 10.7723867893219
  timestamp: 1658500552
  timesteps_since_restore: 0
  training_iteration: 100
  trial_id: 8e30a596
  warmup_time: 0.004169940948486328
  
Result for objective_96417df0:
  date: 2022-07-22_15-35-54
  done: false
  experiment_id: e98b245717b4423d8917589cf5d42088
  hostname: Kais-MacBook-Pro.local
  iterations: 0
  iterations_since_restore: 1
  mean_loss: 10.284
  neg_mean_loss: -10.284
  node_ip: 127.0.0.1
  pid: 47690
  time_since_restore: 0.10359907150268555
  time_this_iter_s: 0.10359907150268555
  time_total_s: 0.10359907150268555
  timestamp: 1658500554
  timesteps_since_restore: 0
  training_iteration: 1
  trial_id: 96417df0
  warmup_time: 0.003325939178466797
  
Result for objective_96435da0:
  date: 2022-07-22_15-35-54
  done: false
  experiment_id: 8680a80a099c452dadd3be44d3457ca5
  hostname: Kais-MacBook-Pro.local
  iterations: 0
  iterations_since_restore: 1
  mean_loss: 10.26
  neg_mean_loss: -10.26
  node_ip: 127.0.0.1
  pid: 47691
  time_since_restore: 0.10206389427185059
  time_this_iter_s: 0.10206389427185059
  time_total_s: 0.10206389427185059
  timestamp: 1658500554
  timesteps_since_restore: 0
  training_iteration: 1
  trial_id: 96435da0
  warmup_time: 0.002891063690185547
  
Result for objective_963faf2a:
  date: 2022-07-22_15-35-54
  done: false
  experiment_id: f28968c2634348c2ae4b0118cc844687
  hostname: Kais-MacBook-Pro.local
  iterations: 0
  iterations_since_restore: 1
  mean_loss: 9.436
  neg_mean_loss: -9.436
  node_ip: 127.0.0.1
  pid: 47689
  time_since_restore: 0.10424089431762695
  time_this_iter_s: 0.10424089431762695
  time_total_s: 0.10424089431762695
  timestamp: 1658500554
  timesteps_since_restore: 0
  training_iteration: 1
  trial_id: 963faf2a
  warmup_time: 0.002835988998413086
  
Result for objective_96417df0:
  date: 2022-07-22_15-35-59
  done: false
  experiment_id: e98b245717b4423d8917589cf5d42088
  hostname: Kais-MacBook-Pro.local
  iterations: 46
  iterations_since_restore: 47
  mean_loss: 0.6336503496503496
  neg_mean_loss: -0.6336503496503496
  node_ip: 127.0.0.1
  pid: 47690
  time_since_restore: 5.134061098098755
  time_this_iter_s: 0.1091001033782959
  time_total_s: 5.134061098098755
  timestamp: 1658500559
  timesteps_since_restore: 0
  training_iteration: 47
  trial_id: 96417df0
  warmup_time: 0.003325939178466797
  
Result for objective_96435da0:
  date: 2022-07-22_15-35-59
  done: false
  experiment_id: 8680a80a099c452dadd3be44d3457ca5
  hostname: Kais-MacBook-Pro.local
  iterations: 47
  iterations_since_restore: 48
  mean_loss: 0.6024657534246576
  neg_mean_loss: -0.6024657534246576
  node_ip: 127.0.0.1
  pid: 47691
  time_since_restore: 5.1902501583099365
  time_this_iter_s: 0.10941314697265625
  time_total_s: 5.1902501583099365
  timestamp: 1658500559
  timesteps_since_restore: 0
  training_iteration: 48
  trial_id: 96435da0
  warmup_time: 0.002891063690185547
  
Result for objective_963faf2a:
  date: 2022-07-22_15-35-59
  done: false
  experiment_id: f28968c2634348c2ae4b0118cc844687
  hostname: Kais-MacBook-Pro.local
  iterations: 46
  iterations_since_restore: 47
  mean_loss: -0.21434965034965026
  neg_mean_loss: 0.21434965034965026
  node_ip: 127.0.0.1
  pid: 47689
  time_since_restore: 5.137102127075195
  time_this_iter_s: 0.10854196548461914
  time_total_s: 5.137102127075195
  timestamp: 1658500559
  timesteps_since_restore: 0
  training_iteration: 47
  trial_id: 963faf2a
  warmup_time: 0.002835988998413086
  
Result for objective_96417df0:
  date: 2022-07-22_15-36-04
  done: false
  experiment_id: e98b245717b4423d8917589cf5d42088
  hostname: Kais-MacBook-Pro.local
  iterations: 93
  iterations_since_restore: 94
  mean_loss: 0.46005633802816903
  neg_mean_loss: -0.46005633802816903
  node_ip: 127.0.0.1
  pid: 47690
  time_since_restore: 10.175445079803467
  time_this_iter_s: 0.10581207275390625
  time_total_s: 10.175445079803467
  timestamp: 1658500564
  timesteps_since_restore: 0
  training_iteration: 94
  trial_id: 96417df0
  warmup_time: 0.003325939178466797
  
Result for objective_96435da0:
  date: 2022-07-22_15-36-05
  done: false
  experiment_id: 8680a80a099c452dadd3be44d3457ca5
  hostname: Kais-MacBook-Pro.local
  iterations: 94
  iterations_since_restore: 95
  mean_loss: 0.4342160278745645
  neg_mean_loss: -0.4342160278745645
  node_ip: 127.0.0.1
  pid: 47691
  time_since_restore: 10.23218584060669
  time_this_iter_s: 0.10732793807983398
  time_total_s: 10.23218584060669
  timestamp: 1658500565
  timesteps_since_restore: 0
  training_iteration: 95
  trial_id: 96435da0
  warmup_time: 0.002891063690185547
  
Result for objective_963faf2a:
  date: 2022-07-22_15-36-05
  done: false
  experiment_id: f28968c2634348c2ae4b0118cc844687
  hostname: Kais-MacBook-Pro.local
  iterations: 93
  iterations_since_restore: 94
  mean_loss: -0.38794366197183094
  neg_mean_loss: 0.38794366197183094
  node_ip: 127.0.0.1
  pid: 47689
  time_since_restore: 10.181179761886597
  time_this_iter_s: 0.10726284980773926
  time_total_s: 10.181179761886597
  timestamp: 1658500565
  timesteps_since_restore: 0
  training_iteration: 94
  trial_id: 963faf2a
  warmup_time: 0.002835988998413086
  
Result for objective_96417df0:
  date: 2022-07-22_15-36-05
  done: true
  experiment_id: e98b245717b4423d8917589cf5d42088
  experiment_tag: 6_activation=relu_tanh,height=2.8400,steps=100,width=6
  hostname: Kais-MacBook-Pro.local
  iterations: 99
  iterations_since_restore: 100
  mean_loss: 0.44956291390728476
  neg_mean_loss: -0.44956291390728476
  node_ip: 127.0.0.1
  pid: 47690
  time_since_restore: 10.820996046066284
  time_this_iter_s: 0.10588788986206055
  time_total_s: 10.820996046066284
  timestamp: 1658500565
  timesteps_since_restore: 0
  training_iteration: 100
  trial_id: 96417df0
  warmup_time: 0.003325939178466797
  
Result for objective_96435da0:
  date: 2022-07-22_15-36-05
  done: true
  experiment_id: 8680a80a099c452dadd3be44d3457ca5
  experiment_tag: 7_activation=relu_tanh,height=2.6000,steps=100,width=6
  hostname: Kais-MacBook-Pro.local
  iterations: 99
  iterations_since_restore: 100
  mean_loss: 0.4255629139072848
  neg_mean_loss: -0.4255629139072848
  node_ip: 127.0.0.1
  pid: 47691
  time_since_restore: 10.769440174102783
  time_this_iter_s: 0.10849618911743164
  time_total_s: 10.769440174102783
  timestamp: 1658500565
  timesteps_since_restore: 0
  training_iteration: 100
  trial_id: 96435da0
  warmup_time: 0.002891063690185547
  
Result for objective_963faf2a:
  date: 2022-07-22_15-36-05
  done: true
  experiment_id: f28968c2634348c2ae4b0118cc844687
  experiment_tag: 5_activation=relu_tanh,height=-5.6400,steps=100,width=6
  hostname: Kais-MacBook-Pro.local
  iterations: 99
  iterations_since_restore: 100
  mean_loss: -0.39843708609271516
  neg_mean_loss: 0.39843708609271516
  node_ip: 127.0.0.1
  pid: 47689
  time_since_restore: 10.826673984527588
  time_this_iter_s: 0.10800504684448242
  time_total_s: 10.826673984527588
  timestamp: 1658500565
  timesteps_since_restore: 0
  training_iteration: 100
  trial_id: 963faf2a
  warmup_time: 0.002835988998413086

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': 6, 'height': -5.64, 'activation': 'relu, tanh'}

Optional: passing the parameter space into the search algorithm¶

We can also pass the parameter space ourselves in the following formats:

continuous dimensions: (continuous, search_range, precision)
discrete dimensions: (discrete, search_range, has_order)
grid dimensions: (grid, grid_list)

In [12]:

space = {
    "height": (ValueType.CONTINUOUS, [-10, 10], 1e-2),
    "width": (ValueType.DISCRETE, [0, 10], True),
    "layers": (ValueType.GRID, [4, 8, 16])
}

ZOOpt again handles constraining the amount of concurrent trials with "parallel_num".

In [13]:

zoopt_search_config = {
    "parallel_num": 8,
    "metric": "mean_loss",
    "mode": "min"
}
algo = ZOOptSearch(
    algo="Asracos",
    budget=num_samples,
    dim_dict=space,
    **zoopt_search_config
)

This time we pass only "steps" and "activation" to the Tune config because "height" and "width" have been passed into ZOOptSearch to create the search_algo. Again, we run the experiment to "min"imize the "mean_loss" of the objective by searching search_config via algo, num_samples times.

In [14]:

tuner = tune.Tuner(
    objective,
    tune_config=tune.TuneConfig(
        metric="mean_loss",
        mode="min",
        search_alg=algo,
        num_samples=num_samples,
    ),
    param_space={"steps": 100},
)
results = tuner.fit()

== Status ==
Current time: 2022-07-22 15:36:35 (running for 00:00:29.76)
Memory usage on this node: 8.2/16.0 GiB
Using FIFO scheduling algorithm.
Resources requested: 0/16 CPUs, 0/0 GPUs, 0.0/5.39 GiB heap, 0.0/2.0 GiB objects
Current best trial: a7fcf02e with mean_loss=-0.7457524893314368 and parameters={'steps': 100, 'height': -8.88, 'width': 7, 'layers': 16}
Result logdir: /Users/kai/ray_results/objective_2022-07-22_15-36-05
Number of trials: 10/10 (10 TERMINATED)

Trial name	status	loc	height	layers	width	loss	iter	total time (s)	iterations	neg_mean_loss
objective_9e64c92e	TERMINATED	127.0.0.1:47713	-8.08	16	0	9.192	100	10.7118	99	-9.192
objective_9ff31930	TERMINATED	127.0.0.1:47718	0.38	16	7	0.180248	100	10.7315	99	-0.180248
objective_9ff47d0c	TERMINATED	127.0.0.1:47719	5.09	4	10	0.609	100	10.7924	99	-0.609
objective_9ff5c2b6	TERMINATED	127.0.0.1:47720	5.26	16	1	1.44343	100	10.7868	99	-1.44343
objective_a7f414d6	TERMINATED	127.0.0.1:47737	0.38	4	7	0.180248	100	10.7232	99	-0.180248
objective_a7f5c682	TERMINATED	127.0.0.1:47738	-2.38	16	7	-0.0957525	100	10.7337	99	0.0957525
objective_a7f7c162	TERMINATED	127.0.0.1:47739	0.38	8	7	0.180248	100	10.7452	99	-0.180248
objective_a7f96fda	TERMINATED	127.0.0.1:47740	0.38	16	4	0.284305	100	10.7079	99	-0.284305
objective_a7fb1844	TERMINATED	127.0.0.1:47741	0.38	8	7	0.180248	100	10.7157	99	-0.180248
objective_a7fcf02e	TERMINATED	127.0.0.1:47742	-8.88	16	7	-0.745752	100	10.7305	99	0.745752

Result for objective_9e64c92e:
  date: 2022-07-22_15-36-08
  done: false
  experiment_id: fafb7d88360d408286e616de3dcd4407
  hostname: Kais-MacBook-Pro.local
  iterations: 0
  iterations_since_restore: 1
  mean_loss: 9.192
  neg_mean_loss: -9.192
  node_ip: 127.0.0.1
  pid: 47713
  time_since_restore: 0.1042020320892334
  time_this_iter_s: 0.1042020320892334
  time_total_s: 0.1042020320892334
  timestamp: 1658500568
  timesteps_since_restore: 0
  training_iteration: 1
  trial_id: 9e64c92e
  warmup_time: 0.0030999183654785156
  
Result for objective_9ff47d0c:
  date: 2022-07-22_15-36-11
  done: false
  experiment_id: de33a45d0c344d3aa344d8cff20b6c37
  hostname: Kais-MacBook-Pro.local
  iterations: 0
  iterations_since_restore: 1
  mean_loss: 10.509
  neg_mean_loss: -10.509
  node_ip: 127.0.0.1
  pid: 47719
  time_since_restore: 0.10373878479003906
  time_this_iter_s: 0.10373878479003906
  time_total_s: 0.10373878479003906
  timestamp: 1658500571
  timesteps_since_restore: 0
  training_iteration: 1
  trial_id: 9ff47d0c
  warmup_time: 0.003734111785888672
  
Result for objective_9ff31930:
  date: 2022-07-22_15-36-11
  done: false
  experiment_id: 404b066d5e154070b5a3f6f6ef3acb33
  hostname: Kais-MacBook-Pro.local
  iterations: 0
  iterations_since_restore: 1
  mean_loss: 10.038
  neg_mean_loss: -10.038
  node_ip: 127.0.0.1
  pid: 47718
  time_since_restore: 0.10415983200073242
  time_this_iter_s: 0.10415983200073242
  time_total_s: 0.10415983200073242
  timestamp: 1658500571
  timesteps_since_restore: 0
  training_iteration: 1
  trial_id: 9ff31930
  warmup_time: 0.0033788681030273438
  
Result for objective_9ff5c2b6:
  date: 2022-07-22_15-36-11
  done: false
  experiment_id: aeb7cb0b0c7c4f6692d47dcd1fe36462
  hostname: Kais-MacBook-Pro.local
  iterations: 0
  iterations_since_restore: 1
  mean_loss: 10.526
  neg_mean_loss: -10.526
  node_ip: 127.0.0.1
  pid: 47720
  time_since_restore: 0.10468196868896484
  time_this_iter_s: 0.10468196868896484
  time_total_s: 0.10468196868896484
  timestamp: 1658500571
  timesteps_since_restore: 0
  training_iteration: 1
  trial_id: 9ff5c2b6
  warmup_time: 0.0027132034301757812
  
Result for objective_9e64c92e:
  date: 2022-07-22_15-36-13
  done: false
  experiment_id: fafb7d88360d408286e616de3dcd4407
  hostname: Kais-MacBook-Pro.local
  iterations: 47
  iterations_since_restore: 48
  mean_loss: 9.192
  neg_mean_loss: -9.192
  node_ip: 127.0.0.1
  pid: 47713
  time_since_restore: 5.111851215362549
  time_this_iter_s: 0.10771799087524414
  time_total_s: 5.111851215362549
  timestamp: 1658500573
  timesteps_since_restore: 0
  training_iteration: 48
  trial_id: 9e64c92e
  warmup_time: 0.0030999183654785156
  
Result for objective_9ff47d0c:
  date: 2022-07-22_15-36-16
  done: false
  experiment_id: de33a45d0c344d3aa344d8cff20b6c37
  hostname: Kais-MacBook-Pro.local
  iterations: 47
  iterations_since_restore: 48
  mean_loss: 0.7173333333333334
  neg_mean_loss: -0.7173333333333334
  node_ip: 127.0.0.1
  pid: 47719
  time_since_restore: 5.179923057556152
  time_this_iter_s: 0.12629103660583496
  time_total_s: 5.179923057556152
  timestamp: 1658500576
  timesteps_since_restore: 0
  training_iteration: 48
  trial_id: 9ff47d0c
  warmup_time: 0.003734111785888672
  
Result for objective_9ff31930:
  date: 2022-07-22_15-36-16
  done: false
  experiment_id: 404b066d5e154070b5a3f6f6ef3acb33
  hostname: Kais-MacBook-Pro.local
  iterations: 47
  iterations_since_restore: 48
  mean_loss: 0.3329852507374631
  neg_mean_loss: -0.3329852507374631
  node_ip: 127.0.0.1
  pid: 47718
  time_since_restore: 5.155292749404907
  time_this_iter_s: 0.10897374153137207
  time_total_s: 5.155292749404907
  timestamp: 1658500576
  timesteps_since_restore: 0
  training_iteration: 48
  trial_id: 9ff31930
  warmup_time: 0.0033788681030273438
  
Result for objective_9ff5c2b6:
  date: 2022-07-22_15-36-16
  done: false
  experiment_id: aeb7cb0b0c7c4f6692d47dcd1fe36462
  hostname: Kais-MacBook-Pro.local
  iterations: 47
  iterations_since_restore: 48
  mean_loss: 2.2803859649122806
  neg_mean_loss: -2.2803859649122806
  node_ip: 127.0.0.1
  pid: 47720
  time_since_restore: 5.177261829376221
  time_this_iter_s: 0.10725784301757812
  time_total_s: 5.177261829376221
  timestamp: 1658500576
  timesteps_since_restore: 0
  training_iteration: 48
  trial_id: 9ff5c2b6
  warmup_time: 0.0027132034301757812
  
Result for objective_9e64c92e:
  date: 2022-07-22_15-36-18
  done: false
  experiment_id: fafb7d88360d408286e616de3dcd4407
  hostname: Kais-MacBook-Pro.local
  iterations: 94
  iterations_since_restore: 95
  mean_loss: 9.192
  neg_mean_loss: -9.192
  node_ip: 127.0.0.1
  pid: 47713
  time_since_restore: 10.170606136322021
  time_this_iter_s: 0.10434508323669434
  time_total_s: 10.170606136322021
  timestamp: 1658500578
  timesteps_since_restore: 0
  training_iteration: 95
  trial_id: 9e64c92e
  warmup_time: 0.0030999183654785156
  
Result for objective_9e64c92e:
  date: 2022-07-22_15-36-19
  done: true
  experiment_id: fafb7d88360d408286e616de3dcd4407
  experiment_tag: 1_height=-8.0800,layers=16,steps=100,width=0
  hostname: Kais-MacBook-Pro.local
  iterations: 99
  iterations_since_restore: 100
  mean_loss: 9.192
  neg_mean_loss: -9.192
  node_ip: 127.0.0.1
  pid: 47713
  time_since_restore: 10.711803197860718
  time_this_iter_s: 0.10728001594543457
  time_total_s: 10.711803197860718
  timestamp: 1658500579
  timesteps_since_restore: 0
  training_iteration: 100
  trial_id: 9e64c92e
  warmup_time: 0.0030999183654785156
  
Result for objective_9ff31930:
  date: 2022-07-22_15-36-21
  done: false
  experiment_id: 404b066d5e154070b5a3f6f6ef3acb33
  hostname: Kais-MacBook-Pro.local
  iterations: 94
  iterations_since_restore: 95
  mean_loss: 0.18770059880239523
  neg_mean_loss: -0.18770059880239523
  node_ip: 127.0.0.1
  pid: 47718
  time_since_restore: 10.197859048843384
  time_this_iter_s: 0.10710310935974121
  time_total_s: 10.197859048843384
  timestamp: 1658500581
  timesteps_since_restore: 0
  training_iteration: 95
  trial_id: 9ff31930
  warmup_time: 0.0033788681030273438
  
Result for objective_9ff47d0c:
  date: 2022-07-22_15-36-21
  done: false
  experiment_id: de33a45d0c344d3aa344d8cff20b6c37
  hostname: Kais-MacBook-Pro.local
  iterations: 94
  iterations_since_restore: 95
  mean_loss: 0.6142631578947368
  neg_mean_loss: -0.6142631578947368
  node_ip: 127.0.0.1
  pid: 47719
  time_since_restore: 10.256150722503662
  time_this_iter_s: 0.11135077476501465
  time_total_s: 10.256150722503662
  timestamp: 1658500581
  timesteps_since_restore: 0
  training_iteration: 95
  trial_id: 9ff47d0c
  warmup_time: 0.003734111785888672
  
Result for objective_9ff5c2b6:
  date: 2022-07-22_15-36-21
  done: false
  experiment_id: aeb7cb0b0c7c4f6692d47dcd1fe36462
  hostname: Kais-MacBook-Pro.local
  iterations: 94
  iterations_since_restore: 95
  mean_loss: 1.4875384615384615
  neg_mean_loss: -1.4875384615384615
  node_ip: 127.0.0.1
  pid: 47720
  time_since_restore: 10.24931287765503
  time_this_iter_s: 0.10830807685852051
  time_total_s: 10.24931287765503
  timestamp: 1658500581
  timesteps_since_restore: 0
  training_iteration: 95
  trial_id: 9ff5c2b6
  warmup_time: 0.0027132034301757812
  
Result for objective_9ff31930:
  date: 2022-07-22_15-36-21
  done: true
  experiment_id: 404b066d5e154070b5a3f6f6ef3acb33
  experiment_tag: 2_height=0.3800,layers=16,steps=100,width=7
  hostname: Kais-MacBook-Pro.local
  iterations: 99
  iterations_since_restore: 100
  mean_loss: 0.18024751066856332
  neg_mean_loss: -0.18024751066856332
  node_ip: 127.0.0.1
  pid: 47718
  time_since_restore: 10.731510877609253
  time_this_iter_s: 0.1073448657989502
  time_total_s: 10.731510877609253
  timestamp: 1658500581
  timesteps_since_restore: 0
  training_iteration: 100
  trial_id: 9ff31930
  warmup_time: 0.0033788681030273438
  
Result for objective_9ff47d0c:
  date: 2022-07-22_15-36-21
  done: true
  experiment_id: de33a45d0c344d3aa344d8cff20b6c37
  experiment_tag: 3_height=5.0900,layers=4,steps=100,width=10
  hostname: Kais-MacBook-Pro.local
  iterations: 99
  iterations_since_restore: 100
  mean_loss: 0.609
  neg_mean_loss: -0.609
  node_ip: 127.0.0.1
  pid: 47719
  time_since_restore: 10.792427778244019
  time_this_iter_s: 0.10576391220092773
  time_total_s: 10.792427778244019
  timestamp: 1658500581
  timesteps_since_restore: 0
  training_iteration: 100
  trial_id: 9ff47d0c
  warmup_time: 0.003734111785888672
  
Result for objective_9ff5c2b6:
  date: 2022-07-22_15-36-21
  done: true
  experiment_id: aeb7cb0b0c7c4f6692d47dcd1fe36462
  experiment_tag: 4_height=5.2600,layers=16,steps=100,width=1
  hostname: Kais-MacBook-Pro.local
  iterations: 99
  iterations_since_restore: 100
  mean_loss: 1.4434311926605505
  neg_mean_loss: -1.4434311926605505
  node_ip: 127.0.0.1
  pid: 47720
  time_since_restore: 10.786811828613281
  time_this_iter_s: 0.10388994216918945
  time_total_s: 10.786811828613281
  timestamp: 1658500581
  timesteps_since_restore: 0
  training_iteration: 100
  trial_id: 9ff5c2b6
  warmup_time: 0.0027132034301757812
  
Result for objective_a7fb1844:
  date: 2022-07-22_15-36-24
  done: false
  experiment_id: 668080a52a3f4c4fad2258177b1fb755
  hostname: Kais-MacBook-Pro.local
  iterations: 0
  iterations_since_restore: 1
  mean_loss: 10.038
  neg_mean_loss: -10.038
  node_ip: 127.0.0.1
  pid: 47741
  time_since_restore: 0.10408973693847656
  time_this_iter_s: 0.10408973693847656
  time_total_s: 0.10408973693847656
  timestamp: 1658500584
  timesteps_since_restore: 0
  training_iteration: 1
  trial_id: a7fb1844
  warmup_time: 0.004377126693725586
  
Result for objective_a7f5c682:
  date: 2022-07-22_15-36-24
  done: false
  experiment_id: 4449a3ef690c4b76a0cfbf7c4de7df43
  hostname: Kais-MacBook-Pro.local
  iterations: 0
  iterations_since_restore: 1
  mean_loss: 9.762
  neg_mean_loss: -9.762
  node_ip: 127.0.0.1
  pid: 47738
  time_since_restore: 0.10507631301879883
  time_this_iter_s: 0.10507631301879883
  time_total_s: 0.10507631301879883
  timestamp: 1658500584
  timesteps_since_restore: 0
  training_iteration: 1
  trial_id: a7f5c682
  warmup_time: 0.003899097442626953
  
Result for objective_a7f414d6:
  date: 2022-07-22_15-36-24
  done: false
  experiment_id: 20575f602c234dd7a167a93ff353299b
  hostname: Kais-MacBook-Pro.local
  iterations: 0
  iterations_since_restore: 1
  mean_loss: 10.038
  neg_mean_loss: -10.038
  node_ip: 127.0.0.1
  pid: 47737
  time_since_restore: 0.10182499885559082
  time_this_iter_s: 0.10182499885559082
  time_total_s: 0.10182499885559082
  timestamp: 1658500584
  timesteps_since_restore: 0
  training_iteration: 1
  trial_id: a7f414d6
  warmup_time: 0.0046689510345458984
  
Result for objective_a7f7c162:
  date: 2022-07-22_15-36-24
  done: false
  experiment_id: f4b185b58be045f594bf127addecbace
  hostname: Kais-MacBook-Pro.local
  iterations: 0
  iterations_since_restore: 1
  mean_loss: 10.038
  neg_mean_loss: -10.038
  node_ip: 127.0.0.1
  pid: 47739
  time_since_restore: 0.10427188873291016
  time_this_iter_s: 0.10427188873291016
  time_total_s: 0.10427188873291016
  timestamp: 1658500584
  timesteps_since_restore: 0
  training_iteration: 1
  trial_id: a7f7c162
  warmup_time: 0.004554033279418945
  
Result for objective_a7fcf02e:
  date: 2022-07-22_15-36-24
  done: false
  experiment_id: 5646116e3b69493b920c031d27eeb11b
  hostname: Kais-MacBook-Pro.local
  iterations: 0
  iterations_since_restore: 1
  mean_loss: 9.112
  neg_mean_loss: -9.112
  node_ip: 127.0.0.1
  pid: 47742
  time_since_restore: 0.10358119010925293
  time_this_iter_s: 0.10358119010925293
  time_total_s: 0.10358119010925293
  timestamp: 1658500584
  timesteps_since_restore: 0
  training_iteration: 1
  trial_id: a7fcf02e
  warmup_time: 0.003072977066040039
  
Result for objective_a7f96fda:
  date: 2022-07-22_15-36-25
  done: false
  experiment_id: 1983d487678f485a8a8ae7da4b7495a0
  hostname: Kais-MacBook-Pro.local
  iterations: 0
  iterations_since_restore: 1
  mean_loss: 10.038
  neg_mean_loss: -10.038
  node_ip: 127.0.0.1
  pid: 47740
  time_since_restore: 0.10406088829040527
  time_this_iter_s: 0.10406088829040527
  time_total_s: 0.10406088829040527
  timestamp: 1658500585
  timesteps_since_restore: 0
  training_iteration: 1
  trial_id: a7f96fda
  warmup_time: 0.0031909942626953125
  
Result for objective_a7fb1844:
  date: 2022-07-22_15-36-29
  done: false
  experiment_id: 668080a52a3f4c4fad2258177b1fb755
  hostname: Kais-MacBook-Pro.local
  iterations: 47
  iterations_since_restore: 48
  mean_loss: 0.3329852507374631
  neg_mean_loss: -0.3329852507374631
  node_ip: 127.0.0.1
  pid: 47741
  time_since_restore: 5.134153842926025
  time_this_iter_s: 0.10844898223876953
  time_total_s: 5.134153842926025
  timestamp: 1658500589
  timesteps_since_restore: 0
  training_iteration: 48
  trial_id: a7fb1844
  warmup_time: 0.004377126693725586
  
Result for objective_a7f5c682:
  date: 2022-07-22_15-36-29
  done: false
  experiment_id: 4449a3ef690c4b76a0cfbf7c4de7df43
  hostname: Kais-MacBook-Pro.local
  iterations: 47
  iterations_since_restore: 48
  mean_loss: 0.05698525073746313
  neg_mean_loss: -0.05698525073746313
  node_ip: 127.0.0.1
  pid: 47738
  time_since_restore: 5.151050090789795
  time_this_iter_s: 0.10810613632202148
  time_total_s: 5.151050090789795
  timestamp: 1658500589
  timesteps_since_restore: 0
  training_iteration: 48
  trial_id: a7f5c682
  warmup_time: 0.003899097442626953
  
Result for objective_a7fcf02e:
  date: 2022-07-22_15-36-29
  done: false
  experiment_id: 5646116e3b69493b920c031d27eeb11b
  hostname: Kais-MacBook-Pro.local
  iterations: 47
  iterations_since_restore: 48
  mean_loss: -0.593014749262537
  neg_mean_loss: 0.593014749262537
  node_ip: 127.0.0.1
  pid: 47742
  time_since_restore: 5.1300742626190186
  time_this_iter_s: 0.10815715789794922
  time_total_s: 5.1300742626190186
  timestamp: 1658500589
  timesteps_since_restore: 0
  training_iteration: 48
  trial_id: a7fcf02e
  warmup_time: 0.003072977066040039
  
Result for objective_a7f414d6:
  date: 2022-07-22_15-36-29
  done: false
  experiment_id: 20575f602c234dd7a167a93ff353299b
  hostname: Kais-MacBook-Pro.local
  iterations: 47
  iterations_since_restore: 48
  mean_loss: 0.3329852507374631
  neg_mean_loss: -0.3329852507374631
  node_ip: 127.0.0.1
  pid: 47737
  time_since_restore: 5.151458024978638
  time_this_iter_s: 0.10657620429992676
  time_total_s: 5.151458024978638
  timestamp: 1658500589
  timesteps_since_restore: 0
  training_iteration: 48
  trial_id: a7f414d6
  warmup_time: 0.0046689510345458984
  
Result for objective_a7f7c162:
  date: 2022-07-22_15-36-29
  done: false
  experiment_id: f4b185b58be045f594bf127addecbace
  hostname: Kais-MacBook-Pro.local
  iterations: 47
  iterations_since_restore: 48
  mean_loss: 0.3329852507374631
  neg_mean_loss: -0.3329852507374631
  node_ip: 127.0.0.1
  pid: 47739
  time_since_restore: 5.154465675354004
  time_this_iter_s: 0.1063377857208252
  time_total_s: 5.154465675354004
  timestamp: 1658500589
  timesteps_since_restore: 0
  training_iteration: 48
  trial_id: a7f7c162
  warmup_time: 0.004554033279418945
  
Result for objective_a7f96fda:
  date: 2022-07-22_15-36-30
  done: false
  experiment_id: 1983d487678f485a8a8ae7da4b7495a0
  hostname: Kais-MacBook-Pro.local
  iterations: 47
  iterations_since_restore: 48
  mean_loss: 0.5430505050505051
  neg_mean_loss: -0.5430505050505051
  node_ip: 127.0.0.1
  pid: 47740
  time_since_restore: 5.135071039199829
  time_this_iter_s: 0.10689830780029297
  time_total_s: 5.135071039199829
  timestamp: 1658500590
  timesteps_since_restore: 0
  training_iteration: 48
  trial_id: a7f96fda
  warmup_time: 0.0031909942626953125
  
Result for objective_a7fb1844:
  date: 2022-07-22_15-36-35
  done: false
  experiment_id: 668080a52a3f4c4fad2258177b1fb755
  hostname: Kais-MacBook-Pro.local
  iterations: 94
  iterations_since_restore: 95
  mean_loss: 0.18770059880239523
  neg_mean_loss: -0.18770059880239523
  node_ip: 127.0.0.1
  pid: 47741
  time_since_restore: 10.180493831634521
  time_this_iter_s: 0.10809183120727539
  time_total_s: 10.180493831634521
  timestamp: 1658500595
  timesteps_since_restore: 0
  training_iteration: 95
  trial_id: a7fb1844
  warmup_time: 0.004377126693725586
  
Result for objective_a7f414d6:
  date: 2022-07-22_15-36-35
  done: false
  experiment_id: 20575f602c234dd7a167a93ff353299b
  hostname: Kais-MacBook-Pro.local
  iterations: 94
  iterations_since_restore: 95
  mean_loss: 0.18770059880239523
  neg_mean_loss: -0.18770059880239523
  node_ip: 127.0.0.1
  pid: 47737
  time_since_restore: 10.188506126403809
  time_this_iter_s: 0.10684013366699219
  time_total_s: 10.188506126403809
  timestamp: 1658500595
  timesteps_since_restore: 0
  training_iteration: 95
  trial_id: a7f414d6
  warmup_time: 0.0046689510345458984
  
Result for objective_a7f5c682:
  date: 2022-07-22_15-36-35
  done: false
  experiment_id: 4449a3ef690c4b76a0cfbf7c4de7df43
  hostname: Kais-MacBook-Pro.local
  iterations: 94
  iterations_since_restore: 95
  mean_loss: -0.08829940119760477
  neg_mean_loss: 0.08829940119760477
  node_ip: 127.0.0.1
  pid: 47738
  time_since_restore: 10.20038390159607
  time_this_iter_s: 0.10745882987976074
  time_total_s: 10.20038390159607
  timestamp: 1658500595
  timesteps_since_restore: 0
  training_iteration: 95
  trial_id: a7f5c682
  warmup_time: 0.003899097442626953
  
Result for objective_a7f7c162:
  date: 2022-07-22_15-36-35
  done: false
  experiment_id: f4b185b58be045f594bf127addecbace
  hostname: Kais-MacBook-Pro.local
  iterations: 94
  iterations_since_restore: 95
  mean_loss: 0.18770059880239523
  neg_mean_loss: -0.18770059880239523
  node_ip: 127.0.0.1
  pid: 47739
  time_since_restore: 10.203772783279419
  time_this_iter_s: 0.1077718734741211
  time_total_s: 10.203772783279419
  timestamp: 1658500595
  timesteps_since_restore: 0
  training_iteration: 95
  trial_id: a7f7c162
  warmup_time: 0.004554033279418945
  
Result for objective_a7fcf02e:
  date: 2022-07-22_15-36-35
  done: false
  experiment_id: 5646116e3b69493b920c031d27eeb11b
  hostname: Kais-MacBook-Pro.local
  iterations: 94
  iterations_since_restore: 95
  mean_loss: -0.738299401197605
  neg_mean_loss: 0.738299401197605
  node_ip: 127.0.0.1
  pid: 47742
  time_since_restore: 10.187027215957642
  time_this_iter_s: 0.10881304740905762
  time_total_s: 10.187027215957642
  timestamp: 1658500595
  timesteps_since_restore: 0
  training_iteration: 95
  trial_id: a7fcf02e
  warmup_time: 0.003072977066040039
  
Result for objective_a7f96fda:
  date: 2022-07-22_15-36-35
  done: false
  experiment_id: 1983d487678f485a8a8ae7da4b7495a0
  hostname: Kais-MacBook-Pro.local
  iterations: 94
  iterations_since_restore: 95
  mean_loss: 0.29706735751295343
  neg_mean_loss: -0.29706735751295343
  node_ip: 127.0.0.1
  pid: 47740
  time_since_restore: 10.170713901519775
  time_this_iter_s: 0.10522603988647461
  time_total_s: 10.170713901519775
  timestamp: 1658500595
  timesteps_since_restore: 0
  training_iteration: 95
  trial_id: a7f96fda
  warmup_time: 0.0031909942626953125
  
Result for objective_a7fb1844:
  date: 2022-07-22_15-36-35
  done: true
  experiment_id: 668080a52a3f4c4fad2258177b1fb755
  experiment_tag: 9_height=0.3800,layers=8,steps=100,width=7
  hostname: Kais-MacBook-Pro.local
  iterations: 99
  iterations_since_restore: 100
  mean_loss: 0.18024751066856332
  neg_mean_loss: -0.18024751066856332
  node_ip: 127.0.0.1
  pid: 47741
  time_since_restore: 10.715673923492432
  time_this_iter_s: 0.10643696784973145
  time_total_s: 10.715673923492432
  timestamp: 1658500595
  timesteps_since_restore: 0
  training_iteration: 100
  trial_id: a7fb1844
  warmup_time: 0.004377126693725586
  
Result for objective_a7f414d6:
  date: 2022-07-22_15-36-35
  done: true
  experiment_id: 20575f602c234dd7a167a93ff353299b
  experiment_tag: 5_height=0.3800,layers=4,steps=100,width=7
  hostname: Kais-MacBook-Pro.local
  iterations: 99
  iterations_since_restore: 100
  mean_loss: 0.18024751066856332
  neg_mean_loss: -0.18024751066856332
  node_ip: 127.0.0.1
  pid: 47737
  time_since_restore: 10.723177909851074
  time_this_iter_s: 0.1052548885345459
  time_total_s: 10.723177909851074
  timestamp: 1658500595
  timesteps_since_restore: 0
  training_iteration: 100
  trial_id: a7f414d6
  warmup_time: 0.0046689510345458984
  
Result for objective_a7f5c682:
  date: 2022-07-22_15-36-35
  done: true
  experiment_id: 4449a3ef690c4b76a0cfbf7c4de7df43
  experiment_tag: 6_height=-2.3800,layers=16,steps=100,width=7
  hostname: Kais-MacBook-Pro.local
  iterations: 99
  iterations_since_restore: 100
  mean_loss: -0.09575248933143668
  neg_mean_loss: 0.09575248933143668
  node_ip: 127.0.0.1
  pid: 47738
  time_since_restore: 10.733658075332642
  time_this_iter_s: 0.10680294036865234
  time_total_s: 10.733658075332642
  timestamp: 1658500595
  timesteps_since_restore: 0
  training_iteration: 100
  trial_id: a7f5c682
  warmup_time: 0.003899097442626953
  
Result for objective_a7f7c162:
  date: 2022-07-22_15-36-35
  done: true
  experiment_id: f4b185b58be045f594bf127addecbace
  experiment_tag: 7_height=0.3800,layers=8,steps=100,width=7
  hostname: Kais-MacBook-Pro.local
  iterations: 99
  iterations_since_restore: 100
  mean_loss: 0.18024751066856332
  neg_mean_loss: -0.18024751066856332
  node_ip: 127.0.0.1
  pid: 47739
  time_since_restore: 10.745207786560059
  time_this_iter_s: 0.10880804061889648
  time_total_s: 10.745207786560059
  timestamp: 1658500595
  timesteps_since_restore: 0
  training_iteration: 100
  trial_id: a7f7c162
  warmup_time: 0.004554033279418945
  
Result for objective_a7fcf02e:
  date: 2022-07-22_15-36-35
  done: true
  experiment_id: 5646116e3b69493b920c031d27eeb11b
  experiment_tag: 10_height=-8.8800,layers=16,steps=100,width=7
  hostname: Kais-MacBook-Pro.local
  iterations: 99
  iterations_since_restore: 100
  mean_loss: -0.7457524893314368
  neg_mean_loss: 0.7457524893314368
  node_ip: 127.0.0.1
  pid: 47742
  time_since_restore: 10.730549097061157
  time_this_iter_s: 0.11018085479736328
  time_total_s: 10.730549097061157
  timestamp: 1658500595
  timesteps_since_restore: 0
  training_iteration: 100
  trial_id: a7fcf02e
  warmup_time: 0.003072977066040039
  
Result for objective_a7f96fda:
  date: 2022-07-22_15-36-35
  done: true
  experiment_id: 1983d487678f485a8a8ae7da4b7495a0
  experiment_tag: 8_height=0.3800,layers=16,steps=100,width=4
  hostname: Kais-MacBook-Pro.local
  iterations: 99
  iterations_since_restore: 100
  mean_loss: 0.2843054187192119
  neg_mean_loss: -0.2843054187192119
  node_ip: 127.0.0.1
  pid: 47740
  time_since_restore: 10.707929849624634
  time_this_iter_s: 0.10735177993774414
  time_total_s: 10.707929849624634
  timestamp: 1658500595
  timesteps_since_restore: 0
  training_iteration: 100
  trial_id: a7f96fda
  warmup_time: 0.0031909942626953125

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

In [15]:

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

Best hyperparameters found were:  {'steps': 100, 'height': -8.88, 'width': 7, 'layers': 16}

In [16]:

ray.shutdown()