[tests] Add tests for Torch PPO (#4429)

ml-agents/mlagents/trainers/tests/torch/test_ppo.py

+import pytest
+
+import numpy as np
+from mlagents.tf_utils import tf
+import copy
+import attr
+
+from mlagents.trainers.ppo.optimizer_torch import TorchPPOOptimizer
+from mlagents.trainers.policy.torch_policy import TorchPolicy
+from mlagents.trainers.tests import mock_brain as mb
+from mlagents.trainers.tests.test_trajectory import make_fake_trajectory
+from mlagents.trainers.settings import NetworkSettings
+from mlagents.trainers.tests.test_simple_rl import PPO_CONFIG
+from mlagents.trainers.tests.test_reward_signals import (  # noqa: F401; pylint: disable=unused-variable
+    curiosity_dummy_config,
+    gail_dummy_config,
+)
+
+
+@pytest.fixture
+def dummy_config():
+    return copy.deepcopy(PPO_CONFIG)
+
+
+VECTOR_ACTION_SPACE = 2
+VECTOR_OBS_SPACE = 8
+DISCRETE_ACTION_SPACE = [3, 3, 3, 2]
+BUFFER_INIT_SAMPLES = 64
+NUM_AGENTS = 12
+
+
+def create_test_ppo_optimizer(dummy_config, use_rnn, use_discrete, use_visual):
+    mock_specs = mb.setup_test_behavior_specs(
+        use_discrete,
+        use_visual,
+        vector_action_space=DISCRETE_ACTION_SPACE
+        if use_discrete
+        else VECTOR_ACTION_SPACE,
+        vector_obs_space=VECTOR_OBS_SPACE,
+    )
+
+    trainer_settings = attr.evolve(dummy_config)
+    trainer_settings.network_settings.memory = (
+        NetworkSettings.MemorySettings(sequence_length=16, memory_size=10)
+        if use_rnn
+        else None
+    )
+    policy = TorchPolicy(0, mock_specs, trainer_settings, "test", False)
+    optimizer = TorchPPOOptimizer(policy, trainer_settings)
+    return optimizer
+
+
+@pytest.mark.parametrize("discrete", [True, False], ids=["discrete", "continuous"])
+@pytest.mark.parametrize("visual", [True, False], ids=["visual", "vector"])
+@pytest.mark.parametrize("rnn", [True, False], ids=["rnn", "no_rnn"])
+def test_ppo_optimizer_update(dummy_config, rnn, visual, discrete):
+    # Test evaluate
+    tf.reset_default_graph()
+    optimizer = create_test_ppo_optimizer(
+        dummy_config, use_rnn=rnn, use_discrete=discrete, use_visual=visual
+    )
+    # Test update
+    update_buffer = mb.simulate_rollout(
+        BUFFER_INIT_SAMPLES,
+        optimizer.policy.behavior_spec,
+        memory_size=optimizer.policy.m_size,
+    )
+    # Mock out reward signal eval
+    update_buffer["advantages"] = update_buffer["environment_rewards"]
+    update_buffer["extrinsic_returns"] = update_buffer["environment_rewards"]
+    update_buffer["extrinsic_value_estimates"] = update_buffer["environment_rewards"]
+
+    # NOTE: In TensorFlow, the log_probs are saved as one for every discrete action, whereas
+    # in PyTorch it is saved as the total probability per branch. So we need to modify the
+    # log prob in the fake buffer here.
+    update_buffer["action_probs"] = np.ones_like(update_buffer["actions"])
+    return_stats = optimizer.update(
+        update_buffer,
+        num_sequences=update_buffer.num_experiences // optimizer.policy.sequence_length,
+    )
+    # Make sure we have the right stats
+    required_stats = [
+        "Losses/Policy Loss",
+        "Losses/Value Loss",
+        "Policy/Learning Rate",
+        "Policy/Epsilon",
+        "Policy/Beta",
+    ]
+    for stat in required_stats:
+        assert stat in return_stats.keys()
+
+
+@pytest.mark.parametrize("discrete", [True, False], ids=["discrete", "continuous"])
+@pytest.mark.parametrize("visual", [True, False], ids=["visual", "vector"])
+@pytest.mark.parametrize("rnn", [True, False], ids=["rnn", "no_rnn"])
+# We need to test this separately from test_reward_signals.py to ensure no interactions
+def test_ppo_optimizer_update_curiosity(
+    dummy_config, curiosity_dummy_config, rnn, visual, discrete  # noqa: F811
+):
+    # Test evaluate
+    tf.reset_default_graph()
+    dummy_config.reward_signals = curiosity_dummy_config
+    optimizer = create_test_ppo_optimizer(
+        dummy_config, use_rnn=rnn, use_discrete=discrete, use_visual=visual
+    )
+    # Test update
+    update_buffer = mb.simulate_rollout(
+        BUFFER_INIT_SAMPLES,
+        optimizer.policy.behavior_spec,
+        memory_size=optimizer.policy.m_size,
+    )
+    # Mock out reward signal eval
+    update_buffer["advantages"] = update_buffer["environment_rewards"]
+    update_buffer["extrinsic_returns"] = update_buffer["environment_rewards"]
+    update_buffer["extrinsic_value_estimates"] = update_buffer["environment_rewards"]
+    update_buffer["curiosity_returns"] = update_buffer["environment_rewards"]
+    update_buffer["curiosity_value_estimates"] = update_buffer["environment_rewards"]
+    # NOTE: In TensorFlow, the log_probs are saved as one for every discrete action, whereas
+    # in PyTorch it is saved as the total probability per branch. So we need to modify the
+    # log prob in the fake buffer here.
+    update_buffer["action_probs"] = np.ones_like(update_buffer["actions"])
+    optimizer.update(
+        update_buffer,
+        num_sequences=update_buffer.num_experiences // optimizer.policy.sequence_length,
+    )
+
+
+# We need to test this separately from test_reward_signals.py to ensure no interactions
+def test_ppo_optimizer_update_gail(gail_dummy_config, dummy_config):  # noqa: F811
+    # Test evaluate
+    dummy_config.reward_signals = gail_dummy_config
+    optimizer = create_test_ppo_optimizer(
+        PPO_CONFIG, use_rnn=False, use_discrete=False, use_visual=False
+    )
+    # Test update
+    update_buffer = mb.simulate_rollout(
+        BUFFER_INIT_SAMPLES, optimizer.policy.behavior_spec
+    )
+    # Mock out reward signal eval
+    update_buffer["advantages"] = update_buffer["environment_rewards"]
+    update_buffer["extrinsic_returns"] = update_buffer["environment_rewards"]
+    update_buffer["extrinsic_value_estimates"] = update_buffer["environment_rewards"]
+    update_buffer["gail_returns"] = update_buffer["environment_rewards"]
+    update_buffer["gail_value_estimates"] = update_buffer["environment_rewards"]
+    optimizer.update(
+        update_buffer,
+        num_sequences=update_buffer.num_experiences // optimizer.policy.sequence_length,
+    )
+
+    # Check if buffer size is too big
+    update_buffer = mb.simulate_rollout(3000, optimizer.policy.behavior_spec)
+    # Mock out reward signal eval
+    update_buffer["advantages"] = update_buffer["environment_rewards"]
+    update_buffer["extrinsic_returns"] = update_buffer["environment_rewards"]
+    update_buffer["extrinsic_value_estimates"] = update_buffer["environment_rewards"]
+    update_buffer["gail_returns"] = update_buffer["environment_rewards"]
+    update_buffer["gail_value_estimates"] = update_buffer["environment_rewards"]
+    # NOTE: In TensorFlow, the log_probs are saved as one for every discrete action, whereas
+    # in PyTorch it is saved as the total probability per branch. So we need to modify the
+    # log prob in the fake buffer here.
+    update_buffer["action_probs"] = np.ones_like(update_buffer["actions"])
+    optimizer.update(
+        update_buffer,
+        num_sequences=update_buffer.num_experiences // optimizer.policy.sequence_length,
+    )
+
+
+@pytest.mark.parametrize("discrete", [True, False], ids=["discrete", "continuous"])
+@pytest.mark.parametrize("visual", [True, False], ids=["visual", "vector"])
+@pytest.mark.parametrize("rnn", [True, False], ids=["rnn", "no_rnn"])
+def test_ppo_get_value_estimates(dummy_config, rnn, visual, discrete):
+    optimizer = create_test_ppo_optimizer(
+        dummy_config, use_rnn=rnn, use_discrete=discrete, use_visual=visual
+    )
+    time_horizon = 15
+    trajectory = make_fake_trajectory(
+        length=time_horizon,
+        observation_shapes=optimizer.policy.behavior_spec.observation_shapes,
+        max_step_complete=True,
+        action_space=DISCRETE_ACTION_SPACE if discrete else VECTOR_ACTION_SPACE,
+        is_discrete=discrete,
+    )
+    run_out, final_value_out = optimizer.get_trajectory_value_estimates(
+        trajectory.to_agentbuffer(), trajectory.next_obs, done=False
+    )
+    for key, val in run_out.items():
+        assert type(key) is str
+        assert len(val) == 15
+
+    run_out, final_value_out = optimizer.get_trajectory_value_estimates(
+        trajectory.to_agentbuffer(), trajectory.next_obs, done=True
+    )
+    for key, val in final_value_out.items():
+        assert type(key) is str
+        assert val == 0.0
+
+    # Check if we ignore terminal states properly
+    optimizer.reward_signals["extrinsic"].use_terminal_states = False
+    run_out, final_value_out = optimizer.get_trajectory_value_estimates(
+        trajectory.to_agentbuffer(), trajectory.next_obs, done=False
+    )
+    for key, val in final_value_out.items():
+        assert type(key) is str
+        assert val != 0.0
+
+
+if __name__ == "__main__":
+    pytest.main()