16a31e2c9fedc654e9117b42b8144adf1d0e4900,examples/reinforcement_learning/tutorial_A3C.py,,,#,259

Before Change

        plt.xlabel("episode")
        plt.ylabel("global running reward")
        plt.savefig("a3c.png")
        plt.show()

        GLOBAL_AC.save_ckpt()

    if args.test:

After Change

    if args.test:
        // ============================= EVALUATION =============================
        GLOBAL_AC.load()
        for episode in range(TEST_EPISODES):
            s = env.reset()
            episode_reward = 0
            while True:
                env.render()
                s = s.astype("float32")  // double to float
                a = GLOBAL_AC.get_action(s, greedy=True)
                s, r, d, _ = env.step(a)
                episode_reward += r
                if d:
                    break
            print(
                "Testing  | Episode: {}/{}  | Episode Reward: {:.4f}  | Running Time: {:.4f}".format(
                    episode + 1, TEST_EPISODES, episode_reward,
                    time.time() - T0))

Instances