Bugfix/ppo mask stats window size (#199)

* fixes issue #198 regarding stats_window_size

* updates changelog

* updates test of stats_window_size

* updates test using maskable env

* removes print statement

docs/misc/changelog.rst

@@ -18,6 +18,7 @@ New Features:
 
 Bug Fixes:
 ^^^^^^^^^^
+- Fixed MaskablePPO ignoring stats_window_size argument
 
 Deprecations:
 ^^^^^^^^^^^^^

sb3_contrib/ppo_mask/ppo_mask.py

@@ -222,8 +222,8 @@ def _setup_learn(
         self.start_time = time.time_ns()
         if self.ep_info_buffer is None or reset_num_timesteps:
             # Initialize buffers if they don't exist, or reinitialize if resetting counters
-            self.ep_info_buffer = deque(maxlen=100)
-            self.ep_success_buffer = deque(maxlen=100)
+            self.ep_info_buffer = deque(maxlen=self._stats_window_size)
+            self.ep_success_buffer = deque(maxlen=self._stats_window_size)
 
         if reset_num_timesteps:
             self.num_timesteps = 0

tests/test_run.py

@@ -150,12 +150,13 @@ def test_advantage_normalization(normalize_advantage):
     model.learn(64)
 
 
-@pytest.mark.parametrize("algo", [TRPO, QRDQN])
+@pytest.mark.parametrize("algo", [TRPO, QRDQN, MaskablePPO])
 @pytest.mark.parametrize("stats_window_size", [1, 42])
 def test_ep_buffers_stats_window_size(algo, stats_window_size):
     """Set stats_window_size for logging to non-default value and check if
     ep_info_buffer and ep_success_buffer are initialized to the correct length"""
-    model = algo("MlpPolicy", "CartPole-v1", stats_window_size=stats_window_size)
+    env = InvalidActionEnvDiscrete() if algo == MaskablePPO else "CartPole-v1"
+    model = algo("MlpPolicy", env, stats_window_size=stats_window_size)
     model.learn(total_timesteps=10)
     assert model.ep_info_buffer.maxlen == stats_window_size
     assert model.ep_success_buffer.maxlen == stats_window_size