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

How can i encode my action in doudizhu? #37

Closed loserZhang closed 4 years ago

loserZhang commented 4 years ago

For example, as a human player, I choose 5553. What is my encoded action in [0, 308]?

AIMan-Zzx commented 4 years ago

555* ?

AIMan-Zzx commented 4 years ago

@loserZhang qq276749553, we can learn together

daochenzha commented 4 years ago

@loserZhang It is a good question. As a human, we may not need to encode actions. The idea of encoding actions is to reduce the action space to make it feasible for learning-based agents. As humans, we don't need any action abstraction. Thus, we may bypass Env and directly pass 5553 to Game.

loserZhang commented 4 years ago

thank you very much------------------ 原始邮件 ------------------ 发件人: "Daochen Zha"notifications@github.com 发送时间: 2019年11月23日(星期六) 凌晨0:43 收件人: "datamllab/rlcard"rlcard@noreply.github.com; 抄送: "loserZhang"1303784270@qq.com;"Mention"mention@noreply.github.com; 主题: Re: [datamllab/rlcard] How can i encode my action in doudizhu? (#37)

@loserZhang It is a good question. As a human, we may not need to encode actions. The idea of encoding actions is to reduce the action space to make it feasible for learning-based agents. As humans, we don't need any action abstraction. Thus, we may bypass Env and directly pass 5553 to Game.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or unsubscribe.