search

datamllab / rlcard

Reinforcement Learning / AI Bots in Card (Poker) Games - Blackjack, Leduc, Texas, DouDizhu, Mahjong, UNO.
http://www.rlcard.org
MIT License
2.86k stars 618 forks source link

老乡，您好。请问怎么获得斗地主的训练数据？ #73

Closed liu6023952 closed 4 years ago

liu6023952 commented 4 years ago

老乡，您好。请问怎么获得斗地主的训练数据？能否发布一个简单demo，实现斗地主人机对战。非常感谢！

daochenzha commented 4 years ago

Hi, @liu6023952 Good questions!

For the training data, you can use run function to obtain data. In the example of https://github.com/datamllab/rlcard/blob/master/examples/doudizhu_random.py, the data are obtained by playing with random agents. We can also generate data with other agents. For example, we have recently implemented a rule agent in dev branch. See https://github.com/datamllab/rlcard/blob/dev/rlcard/models/doudizhu_rule_models.py.

For the demo, we currently support a GUI for replay in another repo https://github.com/datamllab/rlcard-showdown This project is under final testing and will be released soon. But we do not support human-agent GUI for now. We may add this demo to the rlcard-showdown repo in the future.