Пример #1
0
    def test_latent_smoothing(self):
        observation_np = self.env.reset()
        smoothed_policy = RealNVPPolicy(
            input_shapes=self.env.observation_shape,
            output_shape=self.env.action_shape,
            action_range=(
                self.env.action_space.low,
                self.env.action_space.high,
            ),
            hidden_layer_sizes=self.hidden_layer_sizes,
            smoothing_coefficient=0.5,
            observation_keys=self.env.observation_keys,
        )

        np.testing.assert_equal(smoothed_policy._smoothing_x, 0.0)
        self.assertEqual(smoothed_policy._smoothing_alpha, 0.5)
        self.assertEqual(
            smoothed_policy._smoothing_beta,
            np.sqrt((1.0 - np.power(smoothed_policy._smoothing_alpha, 2.0))) /
            (1.0 - smoothed_policy._smoothing_alpha))

        smoothing_x_previous = smoothed_policy._smoothing_x
        for i in range(5):
            action_np = smoothed_policy.action(observation_np).numpy()
            observation_np = self.env.step(action_np)[0]

            self.assertFalse(
                np.all(
                    np.equal(smoothing_x_previous,
                             smoothed_policy._smoothing_x)))
            smoothing_x_previous = smoothed_policy._smoothing_x

        smoothed_policy.reset()

        np.testing.assert_equal(smoothed_policy._smoothing_x, 0.0)
Пример #2
0
    def setUp(self):
        self.env = get_environment('gym', 'Swimmer', 'v3', {})
        self.hidden_layer_sizes = (16, 16)
        self.num_coupling_layers = 2

        self.policy = RealNVPPolicy(
            input_shapes=self.env.observation_shape,
            output_shape=self.env.action_shape,
            action_range=(
                self.env.action_space.low,
                self.env.action_space.high,
            ),
            hidden_layer_sizes=self.hidden_layer_sizes,
            num_coupling_layers=self.num_coupling_layers,
            observation_keys=self.env.observation_keys,
        )
Пример #3
0
class RealNVPPolicyTest(tf.test.TestCase):
    def setUp(self):
        self.env = get_environment('gym', 'Swimmer', 'v3', {})
        self.hidden_layer_sizes = (16, 16)
        self.num_coupling_layers = 2

        self.policy = RealNVPPolicy(
            input_shapes=self.env.observation_shape,
            output_shape=self.env.action_shape,
            action_range=(
                self.env.action_space.low,
                self.env.action_space.high,
            ),
            hidden_layer_sizes=self.hidden_layer_sizes,
            num_coupling_layers=self.num_coupling_layers,
            observation_keys=self.env.observation_keys,
        )

    def test_actions_and_log_probs_symbolic(self):
        observation1_np = self.env.reset()
        observation2_np = self.env.step(self.env.action_space.sample())[0]

        observations_np = type(observation1_np)(
            ((key,
              np.stack((observation1_np[key], observation2_np[key]),
                       axis=0).astype(np.float32))
             for key in observation1_np.keys()))

        observations_tf = tree.map_structure(
            lambda x: tf.constant(x, dtype=x.dtype), observations_np)

        for observations in (observations_tf, observations_np):
            actions = self.policy.actions(observations)
            log_pis = self.policy.log_probs(observations, actions)

            self.assertEqual(actions.shape, (2, *self.env.action_shape))
            self.assertEqual(log_pis.shape, (2, 1))

    def test_actions_and_log_probs_numeric(self):
        observation1_np = self.env.reset()
        observation2_np = self.env.step(self.env.action_space.sample())[0]

        observations_np = type(observation1_np)(
            ((key,
              np.stack((observation1_np[key], observation2_np[key]),
                       axis=0).astype(np.float32))
             for key in observation1_np.keys()))

        actions_np = self.policy.actions(observations_np).numpy()
        log_pis_np = self.policy.log_probs(observations_np, actions_np).numpy()

        self.assertEqual(actions_np.shape, (2, *self.env.action_shape))
        self.assertEqual(log_pis_np.shape, (2, 1))

    def test_env_step_with_actions(self):
        observation_np = self.env.reset()
        action = self.policy.action(observation_np).numpy()
        self.env.step(action)

    def test_trainable_variables(self):
        self.assertEqual(tuple(self.policy.trainable_variables),
                         tuple(self.policy.actions_model.trainable_variables))

        self.assertEqual(tuple(self.policy.trainable_variables),
                         tuple(self.policy.log_pis_model.trainable_variables))

        self.assertEqual(
            tuple(self.policy.trainable_variables),
            tuple(self.policy.deterministic_actions_model.trainable_variables))

        self.assertEqual(
            len(self.policy.trainable_variables),
            self.num_coupling_layers * 2 * (len(self.hidden_layer_sizes) + 1))

    def test_get_diagnostics(self):
        observation1_np = self.env.reset()
        observation2_np = self.env.step(self.env.action_space.sample())[0]

        observations_np = type(observation1_np)(
            ((key,
              np.stack((observation1_np[key], observation2_np[key]),
                       axis=0).astype(np.float32))
             for key in observation1_np.keys()))

        diagnostics = self.policy.get_diagnostics(observations_np)

        self.assertTrue(isinstance(diagnostics, OrderedDict))
        self.assertEqual(tuple(diagnostics.keys()),
                         ('entropy-mean', 'entropy-std', 'raw-actions-mean',
                          'raw-actions-std', 'actions-mean', 'actions-std',
                          'actions-min', 'actions-max'))

        for value in diagnostics.values():
            self.assertTrue(np.isscalar(value))

    def test_serialize_deserialize(self):
        observation1 = self.env.reset()
        observation2 = self.env.step(self.env.action_space.sample())[0]

        observations = type(observation1)(
            ((key, np.stack((observation1[key], observation2[key]),
                            axis=0).astype(np.float32))
             for key in observation1.keys()))

        weights = self.policy.get_weights()
        actions = self.policy.actions(observations).numpy()
        log_pis = self.policy.log_probs(observations, actions).numpy()

        serialized = pickle.dumps(self.policy)
        deserialized = pickle.loads(serialized)

        weights_2 = deserialized.get_weights()
        log_pis_2 = deserialized.log_probs(observations, actions).numpy()

        for weight, weight_2 in zip(weights, weights_2):
            np.testing.assert_array_equal(weight, weight_2)

        np.testing.assert_array_equal(log_pis, log_pis_2)
        np.testing.assert_equal(
            actions.shape,
            deserialized.actions(observations).numpy().shape)

    def test_latent_smoothing(self):
        observation_np = self.env.reset()
        smoothed_policy = RealNVPPolicy(
            input_shapes=self.env.observation_shape,
            output_shape=self.env.action_shape,
            action_range=(
                self.env.action_space.low,
                self.env.action_space.high,
            ),
            hidden_layer_sizes=self.hidden_layer_sizes,
            smoothing_coefficient=0.5,
            observation_keys=self.env.observation_keys,
        )

        np.testing.assert_equal(smoothed_policy._smoothing_x, 0.0)
        self.assertEqual(smoothed_policy._smoothing_alpha, 0.5)
        self.assertEqual(
            smoothed_policy._smoothing_beta,
            np.sqrt((1.0 - np.power(smoothed_policy._smoothing_alpha, 2.0))) /
            (1.0 - smoothed_policy._smoothing_alpha))

        smoothing_x_previous = smoothed_policy._smoothing_x
        for i in range(5):
            action_np = smoothed_policy.action(observation_np).numpy()
            observation_np = self.env.step(action_np)[0]

            self.assertFalse(
                np.all(
                    np.equal(smoothing_x_previous,
                             smoothed_policy._smoothing_x)))
            smoothing_x_previous = smoothed_policy._smoothing_x

        smoothed_policy.reset()

        np.testing.assert_equal(smoothed_policy._smoothing_x, 0.0)