Пример #1
0
 def evaluate_state(self, state: TichuState) -> RewardVector:
     """
     
     :param state: 
     :return: 
     """
     points = state.count_points()
     assert points[0] == points[2] and points[1] == points[3]
     # reward is the difference to the enemy team
     r0 = points[0] - points[1]
     r1 = r0 * -1
     return (r0, r1, r0, r1)
Пример #2
0
 def evaluate_state(self, state: TichuState) -> RewardVector:
     points = state.count_points()
     assert points[0] == points[2] and points[1] == points[3]
     return points