search

starry-sky6688 / MADDPG

Pytorch implementation of the MARL algorithm, MADDPG, which correspondings to the paper "Multi-Agent Actor-Critic for Mixed Cooperative-Competitive Environments".
537 stars 83 forks source link

关于参数问题 #43

Closed MonaHe123 closed 6 months ago

MonaHe123 commented 6 months ago

作者您好，我看您的代码在AC网络定义的时候，定义了max_action，并将其赋值为high_action，我看其他文件中high_action的值为1，那么AC定义文件中的max_action似乎没有改变动作或者输出值。想请教一下为什么要定义一个max_action?

starry-sky6688 commented 6 months ago

max_action和high_action是一个意思，只是为了对网络中输入的action进行归一化

MonaHe123 commented 6 months ago

好嘞好嘞，谢谢您！

-----原始邮件----- 发件人:starry-sky6688 @.> 发送时间:2024-04-15 09:36:27 (星期一) 收件人: starry-sky6688/MADDPG @.> 抄送: MonaHe123 @.>, Author @.> 主题: Re: [starry-sky6688/MADDPG] 关于参数问题 (Issue #43)

max_action和high_action是一个意思，只是为了对网络中输入的action进行归一化

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: @.***>