Fixed racecar observation shape

games/racecar.py

@@ -18,8 +18,8 @@ class MuZeroConfig:
 
 
         ### Game
-        self.observation_shape = (1, 1, 4)  # Dimensions of the game observation, must be 3D (channel, height, width). For a 1D array, please reshape it to (1, 1, length of array)
-        self.action_space = list(range(2))  # Fixed list of all possible actions. You should only edit the length
+        self.observation_shape = (3, 96, 96)  # Dimensions of the game observation, must be 3D (channel, height, width). For a 1D array, please reshape it to (1, 1, length of array)
+        self.action_space = list(range(3))  # Fixed list of all possible actions. You should only edit the length
         self.players = list(range(1))  # List of players. You should only edit the length
         self.stacked_observations = 0  # Number of previous observations and previous actions to add to the current observation
 
@@ -147,8 +147,10 @@ class Game(AbstractGame):
         Returns:
             The new observation, the reward and a boolean if the game has ended.
         """
+        print('action', action)
         observation, reward, done, _ = self.env.step(action)
-        return numpy.array([[observation]]), reward, done
+        observationFormatted = self.formatObservation(observation)
+        return observationFormatted, reward, done
 
     def legal_actions(self):
         """
@@ -170,7 +172,9 @@ class Game(AbstractGame):
         Returns:
             Initial observation of the game.
         """
-        return numpy.array([[self.env.reset()]])
+        initialObservation = numpy.array(self.env.reset())
+        initialObservationFormatted = self.formatObservation(initialObservation)
+        return initialObservationFormatted
 
     def close(self):
         """
@@ -200,3 +204,16 @@ class Game(AbstractGame):
             1: "Push cart to the right",
         }
         return f"{action_number}. {actions[action_number]}"
+    
+    def formatObservation(self, observation):
+        """
+        Formats a game observation to the correct format
+
+        Args:
+            observation: the (96, 96, 3) numpy array of the observation to format
+        
+        Returns:
+            The corresponding (3, 96, 96) numpy array of the observation
+
+        """
+        return numpy.transpose(observation, [2, 0, 1])