dqn.py 文件源码

def __init__(self, env: gym.Env, gamma, epsilon, final_epsilon, final_exp_step):
        """
        :param env: environment
        :param gamma: discount rate
        :param epsilon: initial exploration rate
        :param final_epsilon: final exploration rate
        :param final_exp_step: the step terminating exploration
        """
        self.env = env
        self.action_size = self.env.action_space.n
        self.net = DQN(self.action_size)
        self.target_net = DQN(self.action_size)
        self._gamma = gamma
        self._initial_epsilon = epsilon
        self.epsilon = epsilon
        self._final_epsilon = final_epsilon
        self._final_exp_step = final_exp_step
        if cuda_available:
            self.net.cuda()
            self.target_net.cuda()
        self.update_target_net()