-
########## Evaluation ##########
Timestep: 629402 Average reward is 0.458
########## Evaluation ##########
Timestep: 1258220 Average reward is 0.46
########## Evaluation ##########
Timestep: …
-
how to use the pretrained model, such as nfsp agents. i want to play doudizhu with 3 players, all of them load the pretrained model
-
Python: 3.6.3
MySQL: 8.0.17
你好，我按教程进行操作，进入8080后输入完账号密码，点击注册按钮时，游戏界面红色字体报错：
`pymysql.err.OperationalError: (2003, "Can't connect to MySQL server on'localhost'")`
以下是我的操作记录：
```
E:\Output\…
-
对牌型提示example，新增测试例：
```
cards_rocket = Card.card_ints_from_string('BJ-CJ')
PrettyPrint(list_greater_cards(cards_rocket, cards_candidate))
```
若对方打出火箭，按理来说API不可能打得过，但是API仍然输出：
> card type: bomb
…
-
save the total game tree?
if do that, how to load it.
and the same problem as deep cfr
-
i notice that DQNAgent->_build_model has only two fc net , is it too simple?, and can not get better performance? why not use conv net ?
-
哪位大佬给一下修改choose函数，使choose实现玩家能选择自己出牌的方法
-
-
Traceback (most recent call last):
File "/Library/Python/2.7/site-packages/tornado/web.py", line 1511, in _execute
result = yield result
File "/Library/Python/2.7/site-packages/tornado/gen.…
-
self.db: torndb.Connection = self.application.db
^
提示这里出了问题 setting配置都已更改