コード例 #1
0
ファイル: test_fqe.py プロジェクト: navidmdn/d3rlpy 
def test_fqe(observation_shape, action_size, q_func_factory, scaler,
             action_scaler):
    algo = DDPG()
    fqe = FQE(
        algo=algo,
        scaler=scaler,
        action_scaler=action_scaler,
        q_func_factory=q_func_factory,
    )
    ope_tester(fqe, observation_shape)
    algo.create_impl(observation_shape, action_size)
    algo_update_tester(fqe, observation_shape, action_size, discrete=False)
コード例 #2
0
def test_discrete_fqe(observation_shape, action_size, q_func_factory, scalers):
    scaler, reward_scaler = scalers
    algo = DQN()
    fqe = DiscreteFQE(
        algo=algo,
        scaler=scaler,
        reward_scaler=reward_scaler,
        q_func_factory=q_func_factory,
    )
    ope_tester(fqe, observation_shape)
    algo.create_impl(observation_shape, action_size)
    algo_update_tester(fqe, observation_shape, action_size, discrete=True)