mirror of
https://github.com/saymrwulf/stable-baselines3.git
synced 2026-05-31 23:28:05 +00:00
18d10dbf42
* Update evaluate_policy to use monitor data if available * Update documentation * Cleaning up * Remove unnecessary typing trickery * Update doc * Rename is_wrapped to clarify it is for vecenvs * Add is_wrapped for regular envs * Add is_wrapped call for subprocvecenv and update code for circular imports * Move new functions back to env_util and fix imports * Update changelog * Clarify evaluate_policy docs * Add tests for wrapped modifying episode lengths * Fix tests * Update changelog * Minor edits * Add warn switch to evaluate_policy and update tests Co-authored-by: Antonin RAFFIN <antonin.raffin@ensta.org>
51 lines
1.7 KiB
Python
51 lines
1.7 KiB
Python
import gym
|
|
import numpy as np
|
|
import pytest
|
|
|
|
from stable_baselines3 import DQN, SAC, TD3
|
|
from stable_baselines3.common.evaluation import evaluate_policy
|
|
|
|
|
|
class DummyMultiDiscreteSpace(gym.Env):
|
|
def __init__(self, nvec):
|
|
super(DummyMultiDiscreteSpace, self).__init__()
|
|
self.observation_space = gym.spaces.MultiDiscrete(nvec)
|
|
self.action_space = gym.spaces.Box(low=-1, high=1, shape=(2,), dtype=np.float32)
|
|
|
|
def reset(self):
|
|
return self.observation_space.sample()
|
|
|
|
def step(self, action):
|
|
return self.observation_space.sample(), 0.0, False, {}
|
|
|
|
|
|
class DummyMultiBinary(gym.Env):
|
|
def __init__(self, n):
|
|
super(DummyMultiBinary, self).__init__()
|
|
self.observation_space = gym.spaces.MultiBinary(n)
|
|
self.action_space = gym.spaces.Box(low=-1, high=1, shape=(2,), dtype=np.float32)
|
|
|
|
def reset(self):
|
|
return self.observation_space.sample()
|
|
|
|
def step(self, action):
|
|
return self.observation_space.sample(), 0.0, False, {}
|
|
|
|
|
|
@pytest.mark.parametrize("model_class", [SAC, TD3, DQN])
|
|
@pytest.mark.parametrize("env", [DummyMultiDiscreteSpace([4, 3]), DummyMultiBinary(8)])
|
|
def test_identity_spaces(model_class, env):
|
|
"""
|
|
Additional tests for DQ/SAC/TD3 to check observation space support
|
|
for MultiDiscrete and MultiBinary.
|
|
"""
|
|
# DQN only support discrete actions
|
|
if model_class == DQN:
|
|
env.action_space = gym.spaces.Discrete(4)
|
|
|
|
env = gym.wrappers.TimeLimit(env, max_episode_steps=100)
|
|
|
|
model = model_class("MlpPolicy", env, gamma=0.5, seed=1, policy_kwargs=dict(net_arch=[64]))
|
|
model.learn(total_timesteps=500)
|
|
|
|
evaluate_policy(model, env, n_eval_episodes=5, warn=False)
|