def __init__(self, venv, video_folder, record_video_trigger,
video_length=200, name_prefix='rl-video'):
VecEnvWrapper.__init__(self, venv)
self.env = venv
# Temp variable to retrieve metadata
temp_env = venv
# Unwrap to retrieve metadata dict
# that will be used by gym recorder
while isinstance(temp_env, VecNormalize) or isinstance(temp_env, VecFrameStack):
temp_env = temp_env.venv
if isinstance(temp_env, DummyVecEnv) or isinstance(temp_env, SubprocVecEnv):
metadata = temp_env.get_attr('metadata')[0]
else:
metadata = temp_env.metadata
self.env.metadata = metadata
self.record_video_trigger = record_video_trigger
self.video_recorder = None
self.video_folder = os.path.abspath(video_folder)
# Create output folder if needed
os.makedirs(self.video_folder, exist_ok=True)
self.name_prefix = name_prefix
self.step_id = 0
self.video_length = video_length
self.recording = False
self.recorded_frames = 0
def __init__(self,
venv: VecEnv,
n_stack: int,
channels_order: Optional[Union[str, Dict[str, str]]] = None):
self.venv = venv
self.n_stack = n_stack
wrapped_obs_space = venv.observation_space
if isinstance(wrapped_obs_space, spaces.Box):
assert not isinstance(
channels_order, dict
), f"Expected None or string for channels_order but received {channels_order}"
self.stackedobs = StackedObservations(venv.num_envs, n_stack,
wrapped_obs_space,
channels_order)
elif isinstance(wrapped_obs_space, spaces.Dict):
self.stackedobs = StackedDictObservations(venv.num_envs, n_stack,
wrapped_obs_space,
channels_order)
else:
raise Exception(
"VecFrameStack only works with gym.spaces.Box and gym.spaces.Dict observation spaces"
)
observation_space = self.stackedobs.stack_observation_space(
wrapped_obs_space)
VecEnvWrapper.__init__(self, venv, observation_space=observation_space)
def __init__(
self,
venv: VecEnv,
training: bool = True,
norm_obs: bool = True,
norm_reward: bool = True,
clip_obs: float = 10.0,
clip_reward: float = 10.0,
gamma: float = 0.99,
epsilon: float = 1e-8,
):
VecEnvWrapper.__init__(self, venv)
self.obs_rms = RunningMeanStd(shape=self.observation_space.shape)
self.ret_rms = RunningMeanStd(shape=())
self.clip_obs = clip_obs
self.clip_reward = clip_reward
# Returns: discounted rewards
self.ret = np.zeros(self.num_envs)
self.gamma = gamma
self.epsilon = epsilon
self.training = training
self.norm_obs = norm_obs
self.norm_reward = norm_reward
self.old_obs = np.array([])
self.old_reward = np.array([])
def __init__(self, venv, raise_exception=False, warn_once=True, check_inf=True):
VecEnvWrapper.__init__(self, venv)
self.raise_exception = raise_exception
self.warn_once = warn_once
self.check_inf = check_inf
self._actions = None
self._observations = None
self._user_warned = False
def __init__(self, venv: VecEnv, n_stack: int):
self.venv = venv
self.n_stack = n_stack
wrapped_obs_space = venv.observation_space
low = np.repeat(wrapped_obs_space.low, self.n_stack, axis=-1)
high = np.repeat(wrapped_obs_space.high, self.n_stack, axis=-1)
self.stackedobs = np.zeros((venv.num_envs,) + low.shape, low.dtype)
observation_space = spaces.Box(low=low, high=high, dtype=venv.observation_space.dtype)
VecEnvWrapper.__init__(self, venv, observation_space=observation_space)
def __init__(self, venv: VecEnv, n_stack: int):
self.venv = venv
self.n_stack = n_stack
wrapped_obs_space = venv.observation_space
assert isinstance(
wrapped_obs_space, spaces.Box
), "VecFrameStack only work with gym.spaces.Box observation space"
low = np.repeat(wrapped_obs_space.low, self.n_stack, axis=-1)
high = np.repeat(wrapped_obs_space.high, self.n_stack, axis=-1)
self.stackedobs = np.zeros((venv.num_envs, ) + low.shape, low.dtype)
observation_space = spaces.Box(low=low,
high=high,
dtype=venv.observation_space.dtype)
VecEnvWrapper.__init__(self, venv, observation_space=observation_space)
def set_venv(self, venv: VecEnv) -> None:
"""
Sets the vector environment to wrap to venv.
Also sets attributes derived from this such as `num_env`.
:param venv:
"""
if self.venv is not None:
raise ValueError(
"Trying to set venv of already initialized VecNormalize wrapper."
)
VecEnvWrapper.__init__(self, venv)
if self.obs_rms.mean.shape != self.observation_space.shape:
raise ValueError("venv is incompatible with current statistics.")
self.ret = np.zeros(self.num_envs)
def __init__(
self,
venv: VecEnv,
filename: Optional[str] = None,
info_keywords: Tuple[str, ...] = (),
):
# Avoid circular import
from stable_baselines3.common.monitor import Monitor, ResultsWriter
# This check is not valid for special `VecEnv`
# like the ones created by Procgen, that does follow completely
# the `VecEnv` interface
try:
is_wrapped_with_monitor = venv.env_is_wrapped(Monitor)[0]
except AttributeError:
is_wrapped_with_monitor = False
if is_wrapped_with_monitor:
warnings.warn(
"The environment is already wrapped with a `Monitor` wrapper"
"but you are wrapping it with a `VecMonitor` wrapper, the `Monitor` statistics will be"
"overwritten by the `VecMonitor` ones.",
UserWarning,
)
VecEnvWrapper.__init__(self, venv)
self.episode_returns = None
self.episode_lengths = None
self.episode_count = 0
self.t_start = time.time()
env_id = None
if hasattr(venv, "spec") and venv.spec is not None:
env_id = venv.spec.id
if filename:
self.results_writer = ResultsWriter(filename,
header={
"t_start": self.t_start,
"env_id": env_id
},
extra_keys=info_keywords)
else:
self.results_writer = None
self.info_keywords = info_keywords
def set_venv(self, venv: VecEnv) -> None:
"""
Sets the vector environment to wrap to venv.
Also sets attributes derived from this such as `num_env`.
:param venv:
"""
if self.venv is not None:
raise ValueError(
"Trying to set venv of already initialized VecNormalize wrapper."
)
VecEnvWrapper.__init__(self, venv)
# Check only that the observation_space match
utils.check_for_correct_spaces(venv, self.observation_space,
venv.action_space)
self.returns = np.zeros(self.num_envs)
def __init__(
self,
venv: VecEnv,
training: bool = True,
norm_obs: bool = True,
norm_reward: bool = True,
clip_obs: float = 10.0,
clip_reward: float = 10.0,
gamma: float = 0.99,
epsilon: float = 1e-8,
norm_obs_keys: Optional[List[str]] = None,
):
VecEnvWrapper.__init__(self, venv)
self.norm_obs = norm_obs
self.norm_obs_keys = norm_obs_keys
# Check observation spaces
if self.norm_obs:
self._sanity_checks()
if isinstance(self.observation_space, gym.spaces.Dict):
self.obs_spaces = self.observation_space.spaces
self.obs_rms = {
key: RunningMeanStd(shape=self.obs_spaces[key].shape)
for key in self.norm_obs_keys
}
else:
self.obs_spaces = None
self.obs_rms = RunningMeanStd(
shape=self.observation_space.shape)
self.ret_rms = RunningMeanStd(shape=())
self.clip_obs = clip_obs
self.clip_reward = clip_reward
# Returns: discounted rewards
self.returns = np.zeros(self.num_envs)
self.gamma = gamma
self.epsilon = epsilon
self.training = training
self.norm_obs = norm_obs
self.norm_reward = norm_reward
self.old_obs = np.array([])
self.old_reward = np.array([])
def __init__(
self,
venv: VecEnv,
training: bool = True,
norm_obs: bool = True,
norm_reward: bool = True,
clip_obs: float = 10.0,
clip_reward: float = 10.0,
gamma: float = 0.99,
epsilon: float = 1e-8,
):
VecEnvWrapper.__init__(self, venv)
if norm_obs:
if not isinstance(self.observation_space,
(gym.spaces.Box, gym.spaces.Dict)):
raise ValueError(
"VecNormalize only supports `gym.spaces.Box` and `gym.spaces.Dict` observation spaces"
)
if isinstance(self.observation_space, gym.spaces.Dict):
self.obs_keys = set(self.observation_space.spaces.keys())
self.obs_spaces = self.observation_space.spaces
self.obs_rms = {
key: RunningMeanStd(shape=space.shape)
for key, space in self.obs_spaces.items()
}
else:
self.obs_keys, self.obs_spaces = None, None
self.obs_rms = RunningMeanStd(shape=self.observation_space.shape)
self.ret_rms = RunningMeanStd(shape=())
self.clip_obs = clip_obs
self.clip_reward = clip_reward
# Returns: discounted rewards
self.returns = np.zeros(self.num_envs)
self.gamma = gamma
self.epsilon = epsilon
self.training = training
self.norm_obs = norm_obs
self.norm_reward = norm_reward
self.old_obs = np.array([])
self.old_reward = np.array([])
def __init__(self,
venv: VecEnv,
n_stack: int,
channels_order: Optional[str] = None):
self.venv = venv
self.n_stack = n_stack
wrapped_obs_space = venv.observation_space
assert isinstance(
wrapped_obs_space, spaces.Box
), "VecFrameStack only work with gym.spaces.Box observation space"
if channels_order is None:
# Detect channel location automatically for images
if is_image_space(wrapped_obs_space):
self.channels_first = is_image_space_channels_first(
wrapped_obs_space)
else:
# Default behavior for non-image space, stack on the last axis
self.channels_first = False
else:
assert channels_order in {
"last", "first"
}, "`channels_order` must be one of following: 'last', 'first'"
self.channels_first = channels_order == "first"
# This includes the vec-env dimension (first)
self.stack_dimension = 1 if self.channels_first else -1
repeat_axis = 0 if self.channels_first else -1
low = np.repeat(wrapped_obs_space.low, self.n_stack, axis=repeat_axis)
high = np.repeat(wrapped_obs_space.high,
self.n_stack,
axis=repeat_axis)
self.stackedobs = np.zeros((venv.num_envs, ) + low.shape, low.dtype)
observation_space = spaces.Box(low=low,
high=high,
dtype=venv.observation_space.dtype)
VecEnvWrapper.__init__(self, venv, observation_space=observation_space)