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

out of memory. #2

Open
rookie797 opened this issue Nov 18, 2022 · 12 comments
Open

out of memory. #2

rookie797 opened this issue Nov 18, 2022 · 12 comments

Comments

@rookie797
Copy link

抱歉打扰,在我运行您的代码时,报错runtimeerror: cuda out of memory,我通过降低batch size,换显存更大的GPU运行都无法解决,请问这应该检查什么部分或者如何解决呢?谢谢

@Lei-Kun
Copy link
Owner

Lei-Kun commented Nov 18, 2022 via email

@Lei-Kun
Copy link
Owner

Lei-Kun commented Nov 18, 2022 via email

@weixians
Copy link

weixians commented Dec 5, 2022

我推测,这个问题出在了作者将矩阵邻接矩阵转成了稀疏表示
env_adj = aggr_obs(deepcopy(adj).to(device).to_sparse(), configs.n_j * configs.n_m)

@Lei-Kun
Copy link
Owner

Lei-Kun commented Dec 5, 2022

10 \times10规模是可以在8g 的卡上训练的,30 \times 20 需要大概v100 32g版本的

@Jonas-Nick
Copy link

10 \times10规模是可以在8g 的卡上训练的,30 \times 20 需要大概v100 32g版本的

这个在哪里修改呀

@Jonas-Nick
Copy link

不知道为啥租了个 A100 还是out of memory

@Yihua-Cheng
Copy link

10 \times10规模是可以在8g 的卡上训练的,30 \times 20 需要大概v100 32g版本的

这个在哪里修改呀

在Params里面修改configs的参数

@Xiaomingge666
Copy link

不知道为啥租了个 A100 还是内存不足

我也是,用一张A100显卡还是跑不动

@heyuqiao
Copy link

Params.py文件中,把第7、8行关于算例规模的设置改小就可以跑了,亲测10X5只需要3G

@miemiedexiaoyang
Copy link

Params.py文件中,把第7、8行关于算例规模的设置改小就可以跑了,亲测10X5只需要3G

能分享一下您的代码吗?一直调试报错

@heyuqiao
Copy link

Params.py文件中,把第7、8行关于算例规模的设置改小就可以跑了,亲测10X5只需要3G

能分享一下您的代码吗?一直调试报错

hyq4310,你加我V我发给你吧

@Xiaomingge666
Copy link

已经解决了,把工件和机器数量调小就行

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

8 participants