Source code for catalyst.rl.core.environment

from abc import ABC, abstractmethod

import numpy as np

from gym.spaces import Discrete, Space


class EnvironmentSpec(ABC):
    def __init__(self, visualize=False, mode="train", sampler_id=None):
        self._visualize = visualize
        self._mode = mode
        self._sampler_id = 0 if sampler_id is None else sampler_id

    @property
    def history_len(self) -> int:
        return 1

    @property
    @abstractmethod
    def observation_space(self) -> Space:
        pass

    @property
    @abstractmethod
    def state_space(self) -> Space:
        pass

    @property
    @abstractmethod
    def action_space(self) -> Space:
        pass

    @property
    def reward_space(self) -> Space:
        return Space(shape=(1, ), dtype=np.float32)

    @property
    def discrete_actions(self) -> int:
        return isinstance(self.action_space, Discrete)

    @abstractmethod
    def reset(self):
        pass

    @abstractmethod
    def step(self, action):
        pass