nsq.py 文件源码

def __init__(self, q_function, optimizer,
                 t_max, gamma, i_target, explorer, phi=lambda x: x,
                 average_q_decay=0.999, logger=getLogger(__name__),
                 batch_states=batch_states):

        self.shared_q_function = q_function
        self.target_q_function = copy.deepcopy(q_function)
        self.q_function = copy.deepcopy(self.shared_q_function)

        async.assert_params_not_shared(self.shared_q_function, self.q_function)

        self.optimizer = optimizer

        self.t_max = t_max
        self.gamma = gamma
        self.explorer = explorer
        self.i_target = i_target
        self.phi = phi
        self.logger = logger
        self.average_q_decay = average_q_decay
        self.batch_states = batch_states

        self.t_global = mp.Value('l', 0)
        self.t = 0
        self.t_start = 0
        self.past_action_values = {}
        self.past_states = {}
        self.past_rewards = {}
        self.average_q = 0