예제 #1
0
  def test_no_crash_pendulum(self):
    hparams = trainer_lib.create_hparams(
        "ppo_continuous_action_base",
        TrainTest.test_config)

    hparams.add_hparam("environment_spec", simple_gym_spec("Pendulum-v0"))
    rl_trainer_lib.train(hparams)
예제 #2
0
    def test_no_crash_pendulum(self):
        hparams = trainer_lib.create_hparams("ppo_continuous_action_base",
                                             TrainTest.test_config)

        hparams.add_hparam("environment_spec",
                           rl_models.simple_gym_spec("Pendulum-v0"))
        rl_trainer_lib.train(hparams)
예제 #3
0
    def test_no_crash_cartpole(self):
        hparams = trainer_lib.create_hparams("ppo_discrete_action_base",
                                             TrainTest.test_config)

        hparams.add_hparam("environment_spec",
                           rl_models.simple_gym_spec("CartPole-v0"))
        rl_trainer_lib.train(hparams)