vincentzyx / douzero_for_hlddz_fullauto Goto Github PK

见下图，会卡在底牌现身的地方，不知道这个是不是所说的一定概率无法识别王炸的bug。

进程已结束,退出代码-1073740791 (0xC0000409)

缩放比不为100%时，显示不完整

总是识别不出地主和手牌，但是地主底牌可以，是哪里出了问题

下家 出牌 KKQQJJTT
运行时出现错误，已重置
ValueError("Coordinate 'lower' is less than 'upper'")
Traceback (most recent call last):
File "D:\Study\DouZero_For_HLDDZ_FullAuto\main.py", line 269, in init_cards
self.start()
File "D:\Study\DouZero_For_HLDDZ_FullAuto\main.py", line 525, in start
self.handle_others(self.RPlayedCardsPos, self.RPlayedCard, "下家")
File "D:\Study\DouZero_For_HLDDZ_FullAuto\main.py", line 391, in handle_others
self.waitUntilNoAnimation()
File "D:\Study\DouZero_For_HLDDZ_FullAuto\main.py", line 281, in waitUntilNoAnimation
ani = self.haveAnimation(ms)
File "D:\Study\DouZero_For_HLDDZ_FullAuto\main.py", line 759, in haveAnimation
if not self.compareImage(img.crop(region), lastImg.crop(region)):
File "D:\Programe\py310\lib\site-packages\PIL\Image.py", line 1206, in crop
raise ValueError("Coordinate 'lower' is less than 'upper'")
ValueError: Coordinate 'lower' is less than 'upper'

导出的requrements.txt能不能带上版本号啊, 这**python的包兼容性你又不是不知道...

自动模式下,把把都抢地主,而且每次都超级加倍且明牌,这个是AI通过胜率计算出来的?还是可以通过配置文件配置的...

叫牌预估得分: 0.772 不叫预估得分: 1.234
等待加倍或叫地主...地主牌: DXX22AAAAAKQQQQQJJJTT998877666554444433
Traceback (most recent call last):
File "C:\Gametools\DouZero_For_HLDDZ_FullAuto-main\main.py", line 437, in game_loop
self.before_start()
File "C:\Gametools\DouZero_For_HLDDZ_FullAuto-main\main.py", line 845, in before_start
win_rate = LandlordModel.predict_by_model(cards_str, llcards)
File "C:\Gametools\DouZero_For_HLDDZ_FullAuto-main\LandlordModel.py", line 112, in predict_by_model
action_message, show_action_list = pre_env.step("landlord")
File "C:\Gametools\DouZero_For_HLDDZ_FullAuto-main\douzero\env\game.py", line 156, in step
action, actions_confidence, action_list = self.players[1].act(self.game_infoset)
File "C:\Gametools\DouZero_For_HLDDZ_FullAuto-main\douzero\evaluation\deep_agent.py", line 44, in act
obs = get_obs(infoset, model_type=self.model_type)
File "C:\Gametools\DouZero_For_HLDDZ_FullAuto-main\douzero\env\env.py", line 193, in get_obs
return _get_obs_resnet(infoset, infoset.player_position)
File "C:\Gametools\DouZero_For_HLDDZ_FullAuto-main\douzero\env\env.py", line 598, in _get_obs_resnet
three_landlord_cards = _cards2array(infoset.three_landlord_cards)
File "C:\Gametools\DouZero_For_HLDDZ_FullAuto-main\douzero\env\env.py", line 228, in _cards2array
matrix[:, Card2Column[card]] = NumOnes2Array[num_times]
KeyError: 5

默认为ADP模型，但是训练好的模型路径是否是将_card_play_wp_model_path_中的WP改为ADP就可以

给我安装一下多少钱？

遇到领取奖励和换对手按钮同时显示时（换对手按钮被领取奖励按钮遮罩）
会卡在点击换对手这一步

截图时出现错误: error(1400, 'GetWindowRect', '无效的窗口句柄。')
截图时出现错误: error(1400, 'GetWindowRect', '无效的窗口句柄。')
截图时出现错误: error(1400, 'GetWindowRect', '无效的窗口句柄。')
Traceback (most recent call last):
File "H:\Code Files\Python\Pycharm\DouZero_For_HLDDZ_FullAuto\DouZero_For_HLDDZ_FullAuto-main\debug_screenshot.py", line 7, in
img = gh.DrawRectWithText(img, (313, 747, 1144, 200), "test")
File "H:\Code Files\Python\Pycharm\DouZero_For_HLDDZ_FullAuto\DouZero_For_HLDDZ_FullAuto-main\GameHelper.py", line 29, in DrawRectWithText
img = cv2.cvtColor(image, cv2.COLOR_RGB2BGR)
cv2.error: OpenCV(4.5.5) 👎 error: (-5:Bad argument) in function 'cvtColor'

Overload resolution failed:

• src data type = 17 is not supported
• Expected Ptrcv::UMat for argument 'src'

我猜是 main.py

运行时出现错误，已重置
KeyError(None)
Traceback (most recent call last):
File "C:\Gametools\DouZero_For_HLDDZ_FullAuto-main\main.py", line 269, in init_cards
self.start()
File "C:\Gametools\DouZero_For_HLDDZ_FullAuto-main\main.py", line 525, in start
self.handle_others(self.RPlayedCardsPos, self.RPlayedCard, "下家")
File "C:\Gametools\DouZero_For_HLDDZ_FullAuto-main\main.py", line 379, in handle_others
self.env.step(self.user_position, self.other_played_cards_env)
File "C:\Gametools\DouZero_For_HLDDZ_FullAuto-main\douzero\env\game.py", line 205, in step
self.played_cards[self.acting_player_position] += action
KeyError: None

先使用 debug_screenshot.py 确认自己的屏幕缩放比
请问缩放比是什么 修改哪里？

此页面一点击就报错退出
pygame 2.5.2 (SDL 2.28.3, Python 3.9.18)
Hello from the pygame community. https://www.pygame.org/contribute.html
landlord loads resnet model: baselines/resnet/resnet_landlord.ckpt
现在是自动模式
['-0.2', '-0.1', '0.0', '0.2', '0.1', '0.3', '0.2', '0.3', '0.2', '0.2', '0.1', '0.4']
截图时出现错误: error(5, 'SetActiveWindow', '拒绝访问。')
截图时出现错误: error(5, 'SetActiveWindow', '拒绝访问。')
截图时出现错误: error(5, 'SetActiveWindow', '拒绝访问。')

进程已结束,退出代码-1073740791 (0xC0000409)

我调整了分辨率和缩放比，python环境不存在问题，也在开始游戏的界面点击单局，但是报错：
Process finished with exit code -1073740791 (0xC0000409)
注：我的欢乐斗地主在QQ游戏中心下载，版本 为最新版
尝试了：

• 更改应用的高DPI缩放为系统：仍然报错。

1. 在使用HLDDZ_FallAuto 配置指导最后一步拉起主程序时报错：
   （此时我在PyCharm中运行main.py是可以成功启动的，但是只要点击按钮就会闪退）
   
   报错显示PyQt启动应用程序时，未能将QT平台插件初始化。
   主要有三种可能的错误：

• PyQt5与PyQt5-tool的版本不符合
• 未将QT添加至系统环境变量
• 即使pip命令重装仍然报错，可能需要本地下载相同版本的两个.whl文件重装
  链接：PyQt5, PyQt5-tool
  详细的教程请看：PyQt5解决Qt platform plugin could be initialized

2. 本人的解决方法：

• 检查是否安装了PyQt5与PyQt5-tool
  cmd 输入pip install PyQt5和pip install PyQt5-tool
  发现即使按照requirements.txt的环境安装，但是只会检查是否存在包，不会匹配包的版本，我的两个包的版本不匹配重新安装后解决。
  具体教程可以查看B站视频：https://www.bilibili.com/video/BV1Vq4y1X73o/?vd_source=87ef1f758d28f4e539342e3529e9de8c
• 重新安装后，第一个问题解决，成功拉起主程序
  已经可以自动出牌了。问题解决。

追加bug
我本来以为一切正常，但是再次再PyCharm中运行main.py，有一行警告
DeprecationWarning: sipPyTypeDict() is deprecated, the extension module should use sipPyTypeDictRef() instead class MyPyQT_Form(QtWidgets.QWidget, Ui_Form):
看起来时程序结构的问题，此时UI界面一切正常，但是老毛病又发生了，我点击“单局”按键后闪退。
为此我重新装了pyqt5，pyqt5-plugins，pyqt5-tools，确保不存在多余的版本，且版本之间的相互依赖满足。
仍然无法通过运行main.py运行，但是在FallAuto中正常运行，不太明白发生了什么，但是能运行就好吧。

飞机的画面，很容易卡住

如题，一局结束后如何进入下一局

Cloning into 'Douzero_Cloud_Client'...
remote: Enumerating objects: 726, done.
remote: Counting objects: 100% (33/33), done.
remote: Compressing objects: 100% (6/6), done.
remote: Total 726 (delta 28), reused 27 (delta 27), pack-reused 693
Receiving objects: 100% (726/726), 37.45 MiB | 42.99 MiB/s, done.
Resolving deltas: 100% (420/420), done.
/content/Douzero_Cloud_Client/Douzero_Cloud_Client
Found log directory: douzero_checkpoints/douzero
Saving arguments to douzero_checkpoints/douzero/meta.json
Path to meta file already exists. Not overriding meta.
Saving messages to douzero_checkpoints/douzero/out.log
Path to message file already exists. New data will be appended.
Saving logs data to douzero_checkpoints/douzero/logs.csv
Saving logs' fields to douzero_checkpoints/douzero/fields.csv
Path to log file already exists. New data will be appended.
1
加载模型中，请稍后
初始化，正在获取服务器版本
获取模型版本失败: ConnectionError(MaxRetryError("HTTPConnectionPool(host='cf.vcccz.com', port=80): Max retries exceeded with url: /model_info (Caused by NewConnectionError('<urllib3.connection.HTTPConnection object at 0x7f06fae4ffd0>: Failed to establish a new connection: [Errno -2] Name or service not known'))"))
服务器版本获取失败，更新模型失败
Traceback (most recent call last):
File "/content/Douzero_Cloud_Client/Douzero_Cloud_Client/train.py", line 8, in
train(flags)
File "/content/Douzero_Cloud_Client/Douzero_Cloud_Client/douzero/dmc/dmc.py", line 190, in train
models["cpu"].get_model(position).load_state_dict(torch.load("./models/" + position + ".ckpt", map_location="cpu"))
File "/usr/local/lib/python3.10/dist-packages/torch/serialization.py", line 791, in load
with _open_file_like(f, 'rb') as opened_file:
File "/usr/local/lib/python3.10/dist-packages/torch/serialization.py", line 271, in _open_file_like
return _open_file(name_or_buffer, mode)
File "/usr/local/lib/python3.10/dist-packages/torch/serialization.py", line 252, in init
super().init(open(name, mode))
FileNotFoundError: [Errno 2] No such file or directory: './models/landlord.ckpt'

出现这样的报错
然后就会重新运行

等待下家出牌
下家出牌： 22
等待上家出牌
上家出牌： 3333
File "D:\Project\Python\DouZero_For_HLDDZ_FullAuto-main\mian.py", line 759, in init_cards
self.game_start()
File "D:\Project\Python\DouZero_For_HLDDZ_FullAuto-main\mian.py", line 900, in game_start
action_message, action_list = self.env.step(self.user_position, update=False)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\Project\Python\DouZero_For_HLDDZ_FullAuto-main\douzero\env\game.py", line 157, in step
action, actions_confidence, action_list = self.players[1].act(self.game_infoset)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\Project\Python\DouZero_For_HLDDZ_FullAuto-main\douzero\evaluation\deep_agent.py", line 43, in act
obs = get_obs(infoset, model_type=self.model_type)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\Project\Python\DouZero_For_HLDDZ_FullAuto-main\douzero\env\env.py", line 200, in get_obs
return _get_obs_landlord_down(infoset)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\Project\Python\DouZero_For_HLDDZ_FullAuto-main\douzero\env\env.py", line 500, in _get_obs_landlord_down
landlord_played_cards = _cards2array(
^^^^^^^^^^^^^
File "D:\Project\Python\DouZero_For_HLDDZ_FullAuto-main\douzero\env\env.py", line 228, in _cards2array
matrix[:, Card2Column[card]] = NumOnes2Array[num_times]
~~~~~~~~~~~~~^^^^^^^^^^^
现在是手动模式，请手动出牌
未进入游戏

如题。有时候抢到地主之后程序就卡住了，无法自动进行接下来的加倍出牌等后续操作。