Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

请问用的什么GPU #13

Open
jilner opened this issue Apr 22, 2019 · 9 comments
Open

请问用的什么GPU #13

jilner opened this issue Apr 22, 2019 · 9 comments

Comments

@jilner
Copy link

jilner commented Apr 22, 2019

谢谢作者的工作,我用1080的GPU运行demo都超出内存了,请问用的什么GPU训练和测试呢?谢谢

@Lee-ff
Copy link

Lee-ff commented Sep 8, 2020

谢谢作者的工作,我用1080的GPU运行demo都超出内存了,请问用的什么GPU训练和测试呢?谢谢

你的问题解决了吗

@7568
Copy link

7568 commented Apr 1, 2021

https://github.com/7568/DeRaindrop
这个是我用pytorch根据论文实现的代码,GeForce GTX 1660 Ti 的GPU可以自己训练model

@ChaliceRunRunRun
Copy link

https://github.com/7568/DeRaindrop
这个是我用pytorch根据论文实现的代码,GeForce GTX 1660 Ti的GPU可以自己训练模型
请问pytorch以及其他库的版本有更改吗?

@7568
Copy link

7568 commented Apr 14, 2021

https://github.com/7568/DeRaindrop
这个是我用pytorch根据论文实现的代码,GeForce GTX 1660 Ti的GPU可以自己训练模型
请问pytorch以及其他库的版本有更改吗?

python3.6 其他的都是能兼容python3.6的最新的版本,只要一开始指定了conda的python环境为3.6,其他的包都直接默认安装就行,pip也是一样的,都不需要指定版本

@ChaliceRunRunRun
Copy link

多谢!

@ChaliceRunRunRun
Copy link

https://github.com/7568/DeRaindrop
这个是我用pytorch根据论文实现的代码,GeForce GTX 1660 Ti的GPU可以自己训练模型
请问pytorch以及其他库的版本有更改吗?

python3.6其他的都是兼容python3.6的最新版本,只要一开始指定了conda的python环境为3.6,其他的包都直接替换安装就行,pip也是一样的,都不需要指定版本

您好,最近训练遇到了一个问题,在epoch100轮之前,比如从80轮开始,用这个权重测试,生成的照片就是全黑的,请问是设置问题还是GAN训练不稳定的问题?

@7568
Copy link

7568 commented Apr 21, 2021

之前是有这个问题,是我保存中间训练过程中的模型的时候,少保存了优化器,导致接着训练的时候失败,已经修复了

@ChaliceRunRunRun
Copy link

之前是有这个问题,是我保存中间训练过程中的模型的时候,少保存了优化器,导致接着训练的时候失败,已经修复了

多谢大佬,可以保存权重了。还有一个问题想请教一下,大概训练多少个epoch才能达到和原文相近的效果呢?训练了500轮发现还是和作者给的结果差距很大

@xxx46408
Copy link

@7568 大佬,链接掉了,能重新发一下吗?感谢

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

5 participants