Atari Wrappers

class stable_baselines3.common.atari_wrappers.AtariWrapper(env: gym.core.Env, noop_max: int = 30, frame_skip: int = 4, screen_size: int = 84, terminal_on_life_loss: bool = True, clip_reward: bool = True)

Atari 2600 preprocessings

Specifically:

• NoopReset: obtain initial state by taking random number of no-ops on reset.

• Frame skipping: 4 by default

• Max-pooling: most recent two observations

• Termination signal when a life is lost.

• Resize to a square image: 84x84 by default

• Grayscale observation

• Clip reward to {-1, 0, 1}

Parameters

• env – (gym.Env) gym environment

• noop_max – (int): max number of no-ops

• frame_skip – (int): the frequency at which the agent experiences the game.

• screen_size – (int): resize Atari frame

• terminal_on_life_loss – (bool): if True, then step() returns done=True whenever a life is lost.

• clip_reward – (bool) If True (default), the reward is clip to {-1, 0, 1} depending on its sign.

class stable_baselines3.common.atari_wrappers.ClipRewardEnv(env: gym.core.Env)

reward(reward: float) → float

Bin reward to {+1, 0, -1} by its sign.

Parameters

reward – (float)

Returns

(float)

class stable_baselines3.common.atari_wrappers.EpisodicLifeEnv(env: gym.core.Env)

reset(**kwargs) → numpy.ndarray

Calls the Gym environment reset, only when lives are exhausted. This way all states are still reachable even though lives are episodic, and the learner need not know about any of this behind-the-scenes.

Parameters

kwargs – Extra keywords passed to env.reset() call

Returns

(np.ndarray) the first observation of the environment

step(action: int) → Tuple[Union[Tuple, Dict[str, Any], numpy.ndarray, int], float, bool, Dict]

Run one timestep of the environment’s dynamics. When end of episode is reached, you are responsible for calling reset() to reset this environment’s state.

Accepts an action and returns a tuple (observation, reward, done, info).

Args:

action (object): an action provided by the agent

Returns:

observation (object): agent’s observation of the current environment reward (float) : amount of reward returned after previous action done (bool): whether the episode has ended, in which case further step() calls will return undefined results info (dict): contains auxiliary diagnostic information (helpful for debugging, and sometimes learning)

class stable_baselines3.common.atari_wrappers.FireResetEnv(env: gym.core.Env)

reset(**kwargs) → numpy.ndarray

Resets the state of the environment and returns an initial observation.

Returns:

observation (object): the initial observation.

class stable_baselines3.common.atari_wrappers.MaxAndSkipEnv(env: gym.core.Env, skip: int = 4)

reset(**kwargs)

Resets the state of the environment and returns an initial observation.

Returns:

observation (object): the initial observation.

step(action: int) → Tuple[Union[Tuple, Dict[str, Any], numpy.ndarray, int], float, bool, Dict]

Step the environment with the given action Repeat action, sum reward, and max over last observations.

Parameters

action – ([int] or [float]) the action

Returns

([int] or [float], [float], [bool], dict) observation, reward, done, information

class stable_baselines3.common.atari_wrappers.NoopResetEnv(env: gym.core.Env, noop_max: int = 30)

reset(**kwargs) → numpy.ndarray

Resets the state of the environment and returns an initial observation.

Returns:

observation (object): the initial observation.

class stable_baselines3.common.atari_wrappers.WarpFrame(env: gym.core.Env, width: int = 84, height: int = 84)

observation(frame: numpy.ndarray) → numpy.ndarray

returns the current observation from a frame

Parameters

frame – (np.ndarray) environment frame

Returns

(np.ndarray) the observation