mirror of
https://github.com/Farama-Foundation/Gymnasium.git
synced 2025-08-16 03:38:44 +00:00
9acf9cd367
* add pygame GUI for frozen_lake.py env * add new line at EOF * pre-commit reformat * improve graphics * new images and dynamic window size * darker tile borders and fix ICC profile * pre-commit hook * adjust elf and stool size * Update frozen_lake.py * reformat * fix #2600 * #2600 * add rgb_array support * reformat * test render api change on FrozenLake * add render support for reset on frozenlake * add clock on pygame render * new render api for blackjack * new render api for cliffwalking * new render api for Env class * update reset method, lunar and Env * fix wrapper * fix reset lunar * new render api for box2d envs * new render api for mujoco envs * fix bug * new render api for classic control envs * fix tests * add render_mode None for CartPole * new render api for test fake envs * pre-commit hook * fix FrozenLake * fix FrozenLake * more render_mode to super - frozenlake * remove kwargs from frozen_lake new * pre-commit hook * add deprecated render method * add backwards compatibility * fix test * add _render * move pygame.init() (avoid pygame dependency on init) * fix pygame dependencies * remove collect_render() maintain multi-behaviours .render() * add type hints * fix renderer * don't call .render() with None * improve docstring * add single_rgb_array to all envs * remove None from metadata["render_modes"] * add type hints to test_env_checkers * fix lint * add comments to renderer * add comments to single_depth_array and single_state_pixels * reformat * add deprecation warnings and env.render_mode declaration * fix lint * reformat * fix tests * add docs * fix car racing determinism * remove warning test envs, customizable modes on renderer * remove commments and add todo for env_checker * fix car racing * replace render mode check with assert * update new mujoco * reformat * reformat * change metaclass definition * fix tests * implement mark suggestions (test, docs, sets) * check_render Co-authored-by: J K Terry <jkterry0@gmail.com>
300 lines
10 KiB
Python
300 lines
10 KiB
Python
from multiprocessing import TimeoutError
|
|
|
|
import numpy as np
|
|
import pytest
|
|
|
|
from gym.error import AlreadyPendingCallError, ClosedEnvironmentError, NoAsyncCallError
|
|
from gym.spaces import Box, Discrete, MultiDiscrete, Tuple
|
|
from gym.vector.async_vector_env import AsyncVectorEnv
|
|
from tests.vector.utils import (
|
|
CustomSpace,
|
|
make_custom_space_env,
|
|
make_env,
|
|
make_slow_env,
|
|
)
|
|
|
|
|
|
@pytest.mark.parametrize("shared_memory", [True, False])
|
|
def test_create_async_vector_env(shared_memory):
|
|
env_fns = [make_env("CartPole-v1", i) for i in range(8)]
|
|
try:
|
|
env = AsyncVectorEnv(env_fns, shared_memory=shared_memory)
|
|
finally:
|
|
env.close()
|
|
|
|
assert env.num_envs == 8
|
|
|
|
|
|
@pytest.mark.parametrize("shared_memory", [True, False])
|
|
def test_reset_async_vector_env(shared_memory):
|
|
env_fns = [make_env("CartPole-v1", i) for i in range(8)]
|
|
try:
|
|
env = AsyncVectorEnv(env_fns, shared_memory=shared_memory)
|
|
observations = env.reset()
|
|
finally:
|
|
env.close()
|
|
|
|
assert isinstance(env.observation_space, Box)
|
|
assert isinstance(observations, np.ndarray)
|
|
assert observations.dtype == env.observation_space.dtype
|
|
assert observations.shape == (8,) + env.single_observation_space.shape
|
|
assert observations.shape == env.observation_space.shape
|
|
|
|
try:
|
|
env = AsyncVectorEnv(env_fns, shared_memory=shared_memory)
|
|
observations = env.reset(return_info=False)
|
|
finally:
|
|
env.close()
|
|
|
|
assert isinstance(env.observation_space, Box)
|
|
assert isinstance(observations, np.ndarray)
|
|
assert observations.dtype == env.observation_space.dtype
|
|
assert observations.shape == (8,) + env.single_observation_space.shape
|
|
assert observations.shape == env.observation_space.shape
|
|
|
|
try:
|
|
env = AsyncVectorEnv(env_fns, shared_memory=shared_memory)
|
|
observations, infos = env.reset(return_info=True)
|
|
finally:
|
|
env.close()
|
|
|
|
assert isinstance(env.observation_space, Box)
|
|
assert isinstance(observations, np.ndarray)
|
|
assert observations.dtype == env.observation_space.dtype
|
|
assert observations.shape == (8,) + env.single_observation_space.shape
|
|
assert observations.shape == env.observation_space.shape
|
|
assert isinstance(infos, dict)
|
|
assert all([isinstance(info, dict) for info in infos])
|
|
|
|
|
|
@pytest.mark.parametrize("shared_memory", [True, False])
|
|
@pytest.mark.parametrize("use_single_action_space", [True, False])
|
|
def test_step_async_vector_env(shared_memory, use_single_action_space):
|
|
env_fns = [make_env("CartPole-v1", i) for i in range(8)]
|
|
try:
|
|
env = AsyncVectorEnv(env_fns, shared_memory=shared_memory)
|
|
observations = env.reset()
|
|
|
|
assert isinstance(env.single_action_space, Discrete)
|
|
assert isinstance(env.action_space, MultiDiscrete)
|
|
|
|
if use_single_action_space:
|
|
actions = [env.single_action_space.sample() for _ in range(8)]
|
|
else:
|
|
actions = env.action_space.sample()
|
|
observations, rewards, dones, _ = env.step(actions)
|
|
finally:
|
|
env.close()
|
|
|
|
assert isinstance(env.observation_space, Box)
|
|
assert isinstance(observations, np.ndarray)
|
|
assert observations.dtype == env.observation_space.dtype
|
|
assert observations.shape == (8,) + env.single_observation_space.shape
|
|
assert observations.shape == env.observation_space.shape
|
|
|
|
assert isinstance(rewards, np.ndarray)
|
|
assert isinstance(rewards[0], (float, np.floating))
|
|
assert rewards.ndim == 1
|
|
assert rewards.size == 8
|
|
|
|
assert isinstance(dones, np.ndarray)
|
|
assert dones.dtype == np.bool_
|
|
assert dones.ndim == 1
|
|
assert dones.size == 8
|
|
|
|
|
|
@pytest.mark.parametrize("shared_memory", [True, False])
|
|
def test_call_async_vector_env(shared_memory):
|
|
env_fns = [make_env("CartPole-v1", i, render_mode="rgb_array") for i in range(4)]
|
|
try:
|
|
env = AsyncVectorEnv(env_fns, shared_memory=shared_memory)
|
|
_ = env.reset()
|
|
images = env.call("render")
|
|
gravity = env.call("gravity")
|
|
finally:
|
|
env.close()
|
|
|
|
assert isinstance(images, tuple)
|
|
assert len(images) == 4
|
|
for i in range(4):
|
|
assert len(images[i]) == 1
|
|
assert isinstance(images[i][0], np.ndarray)
|
|
|
|
assert isinstance(gravity, tuple)
|
|
assert len(gravity) == 4
|
|
for i in range(4):
|
|
assert isinstance(gravity[i], float)
|
|
assert gravity[i] == 9.8
|
|
|
|
|
|
@pytest.mark.parametrize("shared_memory", [True, False])
|
|
def test_set_attr_async_vector_env(shared_memory):
|
|
env_fns = [make_env("CartPole-v1", i) for i in range(4)]
|
|
try:
|
|
env = AsyncVectorEnv(env_fns, shared_memory=shared_memory)
|
|
env.set_attr("gravity", [9.81, 3.72, 8.87, 1.62])
|
|
gravity = env.get_attr("gravity")
|
|
assert gravity == (9.81, 3.72, 8.87, 1.62)
|
|
finally:
|
|
env.close()
|
|
|
|
|
|
@pytest.mark.parametrize("shared_memory", [True, False])
|
|
def test_copy_async_vector_env(shared_memory):
|
|
env_fns = [make_env("CartPole-v1", i) for i in range(8)]
|
|
try:
|
|
env = AsyncVectorEnv(env_fns, shared_memory=shared_memory, copy=True)
|
|
observations = env.reset()
|
|
observations[0] = 0
|
|
finally:
|
|
env.close()
|
|
|
|
|
|
@pytest.mark.parametrize("shared_memory", [True, False])
|
|
def test_no_copy_async_vector_env(shared_memory):
|
|
env_fns = [make_env("CartPole-v1", i) for i in range(8)]
|
|
try:
|
|
env = AsyncVectorEnv(env_fns, shared_memory=shared_memory, copy=False)
|
|
observations = env.reset()
|
|
observations[0] = 0
|
|
finally:
|
|
env.close()
|
|
|
|
|
|
@pytest.mark.parametrize("shared_memory", [True, False])
|
|
def test_reset_timeout_async_vector_env(shared_memory):
|
|
env_fns = [make_slow_env(0.3, i) for i in range(4)]
|
|
with pytest.raises(TimeoutError):
|
|
try:
|
|
env = AsyncVectorEnv(env_fns, shared_memory=shared_memory)
|
|
env.reset_async()
|
|
env.reset_wait(timeout=0.1)
|
|
finally:
|
|
env.close(terminate=True)
|
|
|
|
|
|
@pytest.mark.parametrize("shared_memory", [True, False])
|
|
def test_step_timeout_async_vector_env(shared_memory):
|
|
env_fns = [make_slow_env(0.0, i) for i in range(4)]
|
|
with pytest.raises(TimeoutError):
|
|
try:
|
|
env = AsyncVectorEnv(env_fns, shared_memory=shared_memory)
|
|
env.reset()
|
|
env.step_async([0.1, 0.1, 0.3, 0.1])
|
|
observations, rewards, dones, _ = env.step_wait(timeout=0.1)
|
|
finally:
|
|
env.close(terminate=True)
|
|
|
|
|
|
@pytest.mark.filterwarnings("ignore::UserWarning")
|
|
@pytest.mark.parametrize("shared_memory", [True, False])
|
|
def test_reset_out_of_order_async_vector_env(shared_memory):
|
|
env_fns = [make_env("CartPole-v1", i) for i in range(4)]
|
|
with pytest.raises(NoAsyncCallError):
|
|
try:
|
|
env = AsyncVectorEnv(env_fns, shared_memory=shared_memory)
|
|
env.reset_wait()
|
|
except NoAsyncCallError as exception:
|
|
assert exception.name == "reset"
|
|
raise
|
|
finally:
|
|
env.close(terminate=True)
|
|
|
|
with pytest.raises(AlreadyPendingCallError):
|
|
try:
|
|
env = AsyncVectorEnv(env_fns, shared_memory=shared_memory)
|
|
actions = env.action_space.sample()
|
|
env.reset()
|
|
env.step_async(actions)
|
|
env.reset_async()
|
|
except NoAsyncCallError as exception:
|
|
assert exception.name == "step"
|
|
raise
|
|
finally:
|
|
env.close(terminate=True)
|
|
|
|
|
|
@pytest.mark.filterwarnings("ignore::UserWarning")
|
|
@pytest.mark.parametrize("shared_memory", [True, False])
|
|
def test_step_out_of_order_async_vector_env(shared_memory):
|
|
env_fns = [make_env("CartPole-v1", i) for i in range(4)]
|
|
with pytest.raises(NoAsyncCallError):
|
|
try:
|
|
env = AsyncVectorEnv(env_fns, shared_memory=shared_memory)
|
|
actions = env.action_space.sample()
|
|
observations = env.reset()
|
|
observations, rewards, dones, infos = env.step_wait()
|
|
except AlreadyPendingCallError as exception:
|
|
assert exception.name == "step"
|
|
raise
|
|
finally:
|
|
env.close(terminate=True)
|
|
|
|
with pytest.raises(AlreadyPendingCallError):
|
|
try:
|
|
env = AsyncVectorEnv(env_fns, shared_memory=shared_memory)
|
|
actions = env.action_space.sample()
|
|
env.reset_async()
|
|
env.step_async(actions)
|
|
except AlreadyPendingCallError as exception:
|
|
assert exception.name == "reset"
|
|
raise
|
|
finally:
|
|
env.close(terminate=True)
|
|
|
|
|
|
@pytest.mark.parametrize("shared_memory", [True, False])
|
|
def test_already_closed_async_vector_env(shared_memory):
|
|
env_fns = [make_env("CartPole-v1", i) for i in range(4)]
|
|
with pytest.raises(ClosedEnvironmentError):
|
|
env = AsyncVectorEnv(env_fns, shared_memory=shared_memory)
|
|
env.close()
|
|
env.reset()
|
|
|
|
|
|
@pytest.mark.parametrize("shared_memory", [True, False])
|
|
def test_check_spaces_async_vector_env(shared_memory):
|
|
# CartPole-v1 - observation_space: Box(4,), action_space: Discrete(2)
|
|
env_fns = [make_env("CartPole-v1", i) for i in range(8)]
|
|
# FrozenLake-v1 - Discrete(16), action_space: Discrete(4)
|
|
env_fns[1] = make_env("FrozenLake-v1", 1)
|
|
with pytest.raises(RuntimeError):
|
|
env = AsyncVectorEnv(env_fns, shared_memory=shared_memory)
|
|
env.close(terminate=True)
|
|
|
|
|
|
def test_custom_space_async_vector_env():
|
|
env_fns = [make_custom_space_env(i) for i in range(4)]
|
|
try:
|
|
env = AsyncVectorEnv(env_fns, shared_memory=False)
|
|
reset_observations = env.reset()
|
|
|
|
assert isinstance(env.single_action_space, CustomSpace)
|
|
assert isinstance(env.action_space, Tuple)
|
|
|
|
actions = ("action-2", "action-3", "action-5", "action-7")
|
|
step_observations, rewards, dones, _ = env.step(actions)
|
|
finally:
|
|
env.close()
|
|
|
|
assert isinstance(env.single_observation_space, CustomSpace)
|
|
assert isinstance(env.observation_space, Tuple)
|
|
|
|
assert isinstance(reset_observations, tuple)
|
|
assert reset_observations == ("reset", "reset", "reset", "reset")
|
|
|
|
assert isinstance(step_observations, tuple)
|
|
assert step_observations == (
|
|
"step(action-2)",
|
|
"step(action-3)",
|
|
"step(action-5)",
|
|
"step(action-7)",
|
|
)
|
|
|
|
|
|
def test_custom_space_async_vector_env_shared_memory():
|
|
env_fns = [make_custom_space_env(i) for i in range(4)]
|
|
with pytest.raises(ValueError):
|
|
env = AsyncVectorEnv(env_fns, shared_memory=True)
|
|
env.close(terminate=True)
|