last changes

mathema/scape/car_racing.py

@@ -6,6 +6,29 @@ log = logging.getLogger(__name__)
 
 
 class CarRacingScape(Actor):
+    """
+        Scape (environment) actor wrapping a CarRacing-like Gymnasium environment.
+
+        This actor provides an asynchronous message interface for sensors and
+        actuators in the actor-based cortex architecture:
+
+        - Sensors request observations/features via ("sense", sid, sensor_pid).
+          The scape replies to the given sensor actor with ("percept", vec).
+
+        - Actuators apply actions via ("action", action, actuator_pid).
+          The scape performs an env.step(action) and replies with
+          ("result", step_reward, halt_flag) where halt_flag is 1 if the episode
+          terminated or was truncated.
+
+        In addition, the scape automatically resets the environment when an episode
+        ends (halt_flag == 1) using env.fast_reset().
+
+        Notes about `_stepped`:
+        - Some environments do not provide a meaningful feature vector immediately
+          after reset until at least one `step()` was executed.
+        - `_get_features()` ensures that the environment has been stepped once
+          (with a zero action) before calling `env.get_feature_vector()`.
+        """
     def __init__(self, env, name: str = "CarRacingScape"):
         super().__init__(name)
         self.env = env

mathema/scape/scape.py

@@ -1,3 +1,6 @@
+"""
+this is a test scape for validation.
+"""
 from mathema.actors.actor import Actor
 import math
 import logging