ActionWrapper

class maze.core.wrappers.wrapper.ActionWrapper(env: EnvType)

A Wrapper with typing support modifying the agents action.

abstract action(action: Any) → Any

Abstract action mapping method.

get_observation_and_action_dicts(maze_state: Any | None, maze_action: Any | None, first_step_in_episode: bool) → Tuple[Dict[int | str, Any] | None, Dict[int | str, Any] | None]

(overrides Wrapper)

Reverse the actions, keep the observations the same.

abstract reverse_action(action: Any) → Any

Abstract action reverse mapping method.

step(action) → Tuple[Any, Any, bool, Dict[Any, Any]]

(overrides BaseEnv)

Intercept BaseEnv.step and map action.