a3c.py 文件源码

def __init__(self,
                 initial_entropy_beta=0.05,
                 final_entropy_beta=0.0,
                 decay_steps=1e5,
                 thread="global",
                 **settings):

        super(_BaseACNet, self).__init__(**settings)
        self.network_state = None
        self._name_scope = "net_" + str(thread)

        if initial_entropy_beta == final_entropy_beta:
            self._entropy_beta = initial_entropy_beta
        else:
            self._entropy_beta = tf.train.polynomial_decay(
                name="entropy_beta",
                learning_rate=initial_entropy_beta,
                end_learning_rate=final_entropy_beta,
                decay_steps=decay_steps,
                global_step=tf.train.get_global_step())

        with arg_scope([conv2d], data_format="NCHW"), \
             arg_scope([fully_connected, conv2d],
                       activation_fn=self.activation_fn,
                       biases_initializer=tf.constant_initializer(self.init_bias)):
            self.create_architecture()

        self._prepare_loss_op()
        self.params = tf.get_collection(tf.GraphKeys.GLOBAL_VARIABLES, scope=self._name_scope)