修复双ai时ai手牌更新

Author: pineoncellar

douzero/env/game.py

@@ -295,16 +295,15 @@ def get_acting_player_position(self):
 
         return self.acting_player_position
 
-    # 更新手牌
     def update_acting_player_hand_cards(self, action):
         if action != []:
-            # 更新玩家手牌，删除对应的牌
-            if self.acting_player_position == list(self.players.keys())[0]:
+            # 更新AI手牌，删除对应的牌
+            if self.acting_player_position in list(self.players.keys()):
                 for card in action:
                     self.info_sets[
                         self.acting_player_position
                     ].player_hand_cards.remove(card)
-            # 更新另外两个玩家手牌，删除相同数量的牌
+            # 更新玩家手牌，删除相同数量的牌
             else:
                 del self.info_sets[self.acting_player_position].player_hand_cards[
                     0 : len(action)

douzero/evaluation/simulation.py

@@ -180,9 +180,17 @@ def init(data):
 
                 res_data = {
                     "pid": pid,
-                    "cards": cards,
-                    "confidence": confidence,
                     "game_over": res_game_over,
+                    "play": [
+                        {
+                            "cards": cards,
+                            "confidence": confidence,
+                        },
+                        {
+                            "cards": None,
+                            "confidence": None,
+                        },
+                    ],
                 }
                 res_action = "play"
             else:
@@ -223,49 +231,6 @@ def init(data):
                 player_position_code
             ]
 
-            """# 玩家手牌
-            user_hand_cards_real = data["player_data"][0]["hand_cards"]
-            ai0_use_hand_cards_env = [
-                RealCard2EnvCard[c] for c in list(user_hand_cards_real)
-            ]
-            user_hand_cards_real = data["player_data"][1]["hand_cards"]
-            ai1_use_hand_cards_env = [
-                RealCard2EnvCard[c] for c in list(user_hand_cards_real)
-            ]
-            # 角色位号
-            ai0_user_position_code = data["player_data"][0]["position_code"]
-            ai0_user_position = ["landlord_up", "landlord", "landlord_down"][
-                ai0_user_position_code
-            ]
-            ai1_user_position_code = data["player_data"][1]["position_code"]
-            ai1_user_position = ["landlord_up", "landlord", "landlord_down"][
-                ai1_user_position_code
-            ]
-            player_position_code = 3 - ai0_user_position_code - ai1_user_position_code
-            player_position = ["landlord_up", "landlord", "landlord_down"][
-                player_position_code
-            ]
-            # 三张底牌
-            three_landlord_cards_real = data["three_landlord_cards"]
-            three_landlord_cards_env = [
-                RealCard2EnvCard[c] for c in list(three_landlord_cards_real)
-            ]
-            # AI模型
-            ai0_model_name = data["player_data"][0]["model"]
-            ai0_model_list = check_model(ai0_model_name)
-            if ai0_model_list == None:
-                error = Exception(f"找不到此模型：{ai0_model_name}")
-                raise error
-            else:
-                ai0_model = ai0_model_list[ai0_user_position_code]
-            ai1_model_name = data["player_data"][1]["model"]
-            ai1_model_list = check_model(ai1_model_name)
-            if ai1_model_list == None:
-                error = Exception(f"找不到此模型：{ai1_model_name}")
-                raise error
-            else:
-                ai1_model = ai1_model_list[ai1_user_position_code]"""
-
             # 减去两位AI的牌，剩下的牌就是玩家的手牌
             other_hand_cards = []
             for i in set(AllEnvCard):

start.py

@@ -1,6 +1,6 @@
 port = 24813
 ###############################################################################################
-version = 0.8
+version = 1.0
 ###############################################################################################
 import os
 import time