test_discrete.py 文件源码

def test_duel_dqn():
    env = TwoRoundDeterministicRewardEnv()
    np.random.seed(123)
    env.seed(123)
    random.seed(123)
    nb_actions = env.action_space.n

    # Next, we build a very simple model.
    model = Sequential()
    model.add(Dense(16, input_shape=(1,)))
    model.add(Activation('relu'))
    model.add(Dense(nb_actions, activation='linear'))

    memory = SequentialMemory(limit=1000, window_length=1)
    policy = EpsGreedyQPolicy(eps=.1)
    dqn = DQNAgent(model=model, nb_actions=nb_actions, memory=memory, nb_steps_warmup=50,
                   target_model_update=1e-1, policy=policy, enable_double_dqn=False, enable_dueling_network=True)
    dqn.compile(Adam(lr=1e-3))

    dqn.fit(env, nb_steps=2000, visualize=False, verbose=0)
    policy.eps = 0.
    h = dqn.test(env, nb_episodes=20, visualize=False)
    assert_allclose(np.mean(h.history['episode_reward']), 3.)