src.environment.parallelenv.worker.base

Module Contents

Classes

EnvWorker

An abstract worker for an environment.

API

class src.environment.parallelenv.worker.base.EnvWorker(env_fn: Callable[[], gym.Env])

Bases: abc.ABC

An abstract worker for an environment.

Initialization

abstractmethod get_env_attr(key: str) → Any

abstractmethod set_env_attr(key: str, value: Any) → None

send(action: Optional[numpy.ndarray]) → None

Send action signal to low-level worker.

When action is None, it indicates sending “reset” signal; otherwise it indicates “step” signal. The paired return value from “recv” function is determined by such kind of different signal.

send_reset() → None

recv() → Union[Tuple[numpy.ndarray, numpy.ndarray, numpy.ndarray, numpy.ndarray], numpy.ndarray]

Receive result from low-level worker.

If the last “send” function sends a NULL action, it only returns a single observation; otherwise it returns a tuple of (obs, rew, done, info).

reset() → numpy.ndarray

step(action: numpy.ndarray) → Tuple[numpy.ndarray, numpy.ndarray, numpy.ndarray, numpy.ndarray]

Perform one timestep of the environment’s dynamic.

“send” and “recv” are coupled in sync simulation, so users only call “step” function. But they can be called separately in async simulation, i.e. someone calls “send” first, and calls “recv” later.

abstractmethod static wait(workers: List[src.environment.parallelenv.worker.base.EnvWorker], wait_num: int, timeout: Optional[float] = None) → List[src.environment.parallelenv.worker.base.EnvWorker]

Given a list of workers, return those ready ones.

abstractmethod render(**kwargs: Any) → Any

Render the environment.

abstractmethod close_env() → None

close() → None