Source code for verigym.policy.randomized

from copy import deepcopy
from .policy import PolicyClass
from verigym.environments.verigymenv import VeriGymEnv
from verigym.abstraction.abstractionmapper import AbstractionMapper



class RandomizedPolicy(PolicyClass):
    """
    A policy that returns random actions, as sampled from the provided environment.
    Works for every class inheriting from `VeriGymEnv` (and therefore `gym.Env`).
    """

    def __init__(self, env:VeriGymEnv, seed=None):

        ### Create copy of action space to allow for seeded sampling without changing original env.
        self.action_space = deepcopy(env.action_space)
        self.seed(seed)

        def policy(obs):
            return self.action_space.sample()

        abstraction_mapper = AbstractionMapper.initialize_identity_mapper(env.observation_space, env.action_space)  # Identity mapping
        return super().__init__(policy, abstraction_mapper)



    def seed(self, seed=None):
        self.action_space.seed(seed)


    def _action_from_policy(self, obs):
        return self.policy(obs)