Fix 2Rooms Reward

2020-06-30 13:17:11 +09:00 · 2020-06-30 13:17:11 +09:00 · f23b39067a
commit f23b39067a
parent 142b42e34f
1 changed files with 4 additions and 1 deletions
--- a/mujoco_maze/maze_task.py
+++ b/mujoco_maze/maze_task.py
@ -150,7 +150,10 @@ class SingleGoalSparse2Rooms(MazeTask):
        self.goals = [MazeGoal(np.array([0.0, 4.0 * scale]))]

    def reward(self, obs: np.ndarray) -> float:
-        return 1.0 if self.termination(obs) else -0.0001
+        for goal in self.goals:
+            if goal.neighbor(obs):
+                return goal.reward_scale
+        return -0.0001

    @staticmethod
    def create_maze() -> List[List[MazeCell]]: