hyperband_examples.py 2.47 KB
Newer Older
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
#!/usr/bin/env python

import argparse
import json
import os
import random

import numpy as np

import ray
from ray.tune import Trainable, run, Experiment, sample_from
from ray.tune.schedulers import HyperBandScheduler


class MyTrainableClass(Trainable):
    """Example agent whose learning curve is a random sigmoid.

    The dummy hyperparameters "width" and "height" determine the slope and
    maximum reward value reached.
    """

    def _setup(self, config):
        self.timestep = 0

    def _train(self):
        self.timestep += 1
        v = np.tanh(float(self.timestep) / self.config.get("width", 1))
        v *= self.config.get("height", 1)

        # Here we use `episode_reward_mean`, but you can also report other
        # objectives such as loss or accuracy.
        return {"episode_reward_mean": v}

    def _save(self, checkpoint_dir):
        path = os.path.join(checkpoint_dir, "checkpoint")
        with open(path, "w") as f:
            f.write(json.dumps({"timestep": self.timestep}))
        return path

    def _restore(self, checkpoint_path):
        with open(checkpoint_path) as f:
            self.timestep = json.loads(f.read())["timestep"]


if __name__ == "__main__":
    parser = argparse.ArgumentParser()
    parser.add_argument(
        "--smoke-test", action="store_true", help="Finish quickly for testing")
    args, _ = parser.parse_known_args()

    # Hyperband early stopping, configured with `episode_reward_mean` as the
    # objective and `training_iteration` as the time unit,
    # which is automatically filled by Tune.
    hyperband = HyperBandScheduler(
        time_attr="training_iteration",
        metric="episode_reward_mean",
        mode="max",
        max_t=100)

    exp = Experiment(
        name="hyperband_test",
        run=MyTrainableClass,
        num_samples=10,
        stop={"training_iteration": 1 if args.smoke_test else 99999},
        config={
            "width": sample_from(lambda spec: 10 + int(90 * random.random())),
            "height": sample_from(lambda spec: int(100 * random.random()))
        })


    # The reserved lines. The value will be assigned by SLURM
    ray.init(redis_address=os.environ["RAY_HEAD_IP"]) # Assigned by SLURM

    # The local_dir controls where to save the intermediate files. The default place, "~/ray_results", will
    # cause file flush issuses to ray on our cluster.
    exp.spec['local_dir'] = os.environ["Localdir"]
    run(exp, scheduler=hyperband, resources_per_trial={'gpu': os.environ["NUM_GPUs"]})