6ca2a6359dc1374bfb211da8680f3d5f319cdaa5,examples/reinforcement_learning/tutorial_bipedalwalker_a3c_continuous_action.py,,,#,254

Before Change

if __name__ == "__main__":
    sess = tf.Session()

    // ============================= TRAINING ===============================
    with tf.device("/cpu:0"):
        OPT_A = tf.train.RMSPropOptimizer(LR_A, name="RMSPropA")
        OPT_C = tf.train.RMSPropOptimizer(LR_C, name="RMSPropC")
        GLOBAL_AC = ACNet(GLOBAL_NET_SCOPE)  // we only need its params
        workers = []
        // Create worker
        for i in range(N_WORKERS):
            i_name = "Worker_%i" % i  // worker name
            workers.append(Worker(i_name, GLOBAL_AC))

    COORD = tf.train.Coordinator()
    sess.run(tf.global_variables_initializer())

    // start TF threading
    worker_threads = []
    for worker in workers:

After Change

    GLOBAL_AC.load_ckpt()
    while True:
        s = env.reset()
        rall = 0
        while True:
            env.render()
            s = s.astype("float32")  // double to float
            a = GLOBAL_AC.choose_action(s)
            s, r, d, _ = env.step(a)
            rall += r
            if d:
                print("reward", rall)
                break

Instances