Esempio n. 1
0
    def test_train_mode_reset(self):
        """Test reset() in train mode."""
        train_env = Env(mode="train")
        x, x_, theta, theta_ = train_env.reset()

        assert abs(x) <= 2.3
        assert x_ == 0
        assert abs(theta) <= 0.3
        assert theta_ == 0
Esempio n. 2
0
    def test_eval_mode_reset(self):
        """Test reset() in eval mode."""
        eval_env = Env(mode="eval")
        x, x_, theta, theta_ = eval_env.reset()

        assert abs(x) <= 1.0
        assert x_ == 0
        assert abs(theta) <= 0.3
        assert theta_ == 0