mirror of
https://github.com/vale981/ray
synced 2025-03-09 12:56:46 -04:00
60d4d5e1aa
* Remove all __future__ imports from RLlib. * Remove (object) again from tf_run_builder.py::TFRunBuilder. * Fix 2xLINT warnings. * Fix broken appo_policy import (must be appo_tf_policy) * Remove future imports from all other ray files (not just RLlib). * Remove future imports from all other ray files (not just RLlib). * Remove future import blocks that contain `unicode_literals` as well. Revert appo_tf_policy.py to appo_policy.py (belongs to another PR). * Add two empty lines before Schedule class. * Put back __future__ imports into determine_tests_to_run.py. Fails otherwise on a py2/print related error.
38 lines
1.3 KiB
Python
38 lines
1.3 KiB
Python
"""Integration test: (1) pendulum works, (2) single-agent multi-agent works."""
|
|
|
|
import ray
|
|
from ray.rllib.tests.test_multi_agent_env import make_multiagent
|
|
from ray.tune import run_experiments
|
|
from ray.tune.registry import register_env
|
|
|
|
if __name__ == "__main__":
|
|
ray.init()
|
|
MultiPendulum = make_multiagent("Pendulum-v0")
|
|
register_env("multi_pend", lambda _: MultiPendulum(1))
|
|
trials = run_experiments({
|
|
"test": {
|
|
"run": "PPO",
|
|
"env": "multi_pend",
|
|
"stop": {
|
|
"timesteps_total": 500000,
|
|
"episode_reward_mean": -200,
|
|
},
|
|
"config": {
|
|
"train_batch_size": 2048,
|
|
"vf_clip_param": 10.0,
|
|
"num_workers": 0,
|
|
"num_envs_per_worker": 10,
|
|
"lambda": 0.1,
|
|
"gamma": 0.95,
|
|
"lr": 0.0003,
|
|
"sgd_minibatch_size": 64,
|
|
"num_sgd_iter": 10,
|
|
"model": {
|
|
"fcnet_hiddens": [64, 64],
|
|
},
|
|
"batch_mode": "complete_episodes",
|
|
},
|
|
}
|
|
})
|
|
if trials[0].last_result["episode_reward_mean"] < -200:
|
|
raise ValueError("Did not get to -200 reward", trials[0].last_result)
|