driver.py 文件源码

def __init__(self, environment):

        self.env = environment
        self.algorithm = MGAIL(environment=self.env)
        self.init_graph = tf.global_variables_initializer()
        self.saver = tf.train.Saver()
        self.sess = tf.Session()
        if self.env.trained_model:
            self.saver.restore(self.sess, self.env.trained_model)
        else:
            self.sess.run(self.init_graph)
        self.run_dir = self.env.run_dir
        self.loss = 999. * np.ones(3)
        self.reward_mean = 0
        self.reward_std = 0
        self.run_avg = 0.001
        self.discriminator_policy_switch = 0
        self.policy_loop_time = 0
        self.disc_acc = 0
        self.er_count = 0
        self.itr = 0
        self.best_reward = 0
        self.mode = 'Prep'
        np.set_printoptions(precision=2)
        np.set_printoptions(linewidth=220)