rlcard
rlcard copied to clipboard
datamllab
Reinforcement Learning / AI Bots in Card (Poker) Games - Blackjack, Leduc, Texas, DouDizhu, Mahjong, UNO.
I have replaced the DMC running device with CUDA, and the network output is all 0. The output of this action is more like a random selection action. May I...
I am new to this framework and wishing to make Game of 29 AI that will perform bidding and play the hands as well . Any kind of insights or...
如题，有相关的平台吗。 还是说需要自己去把别人的算法都实现一遍，然后把agent放在一起去play，然后比较胜率。 如果要自己去实现也太麻烦了。
Hi I was reading up the RLCard page about Dou Dizhu and it mentioned an action state of 27472 or just 309 by combining reduncencies. I was wondering where self.state_shape...
Trying to run `run_rl.py` pettingzoo example and I get the following error: ``` poetry run python examples/run_rl.py --cuda 0 --> Running on the GPU ---------------------------------------- timestep | 0 reward |...
The paper I read represents the result is very good convergence, but when I train using your code (not change anything), the model not converage and so result's chart is...
牌谱
有考虑给出指定初始信息（手里的牌、上下顺序等）来辅助出牌的计划吗？ Are any plans on supporting AI-assist decision-making with initialized known info?
➜ pip3 install rlcard[torch] zsh: no matches found: rlcard[torch] Platform: macOS with apple silicon; torch and torchvision are installed
看nfsp里，第2个player是random agent, 产生random动作，能像第一个agent一样也用nfsp agent吗，是因为random探索更好更全吗，看训练时只用nfsp agent产生的数据，如果两个player同样用一种 agent, 是不是数据搜集速度翻倍啊。想用在近似无限回合的game, 如果用random探索会不会太慢啊。 非常感谢， ant
