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

<<论文复现>>如何对齐paddle中随机数种子与pytorch中随机数种子 #26637

Closed
xjhcassy opened this issue Aug 25, 2020 · 8 comments
Closed

Comments

@xjhcassy
Copy link

No description provided.

@LielinJiang
Copy link
Contributor

当前版本还无法做到。具体是用在什么层中呢?

@xjhcassy
Copy link
Author

全局的,像torch.manual_seed(opt.manual_seed)这样

@zhiqiu
Copy link
Contributor

zhiqiu commented Aug 25, 2020

全局的,像torch.manual_seed(opt.manual_seed)这样

hi,我们正在支持torch类似的manual_seed和generator功能。

@GuoQuanhao
Copy link

这个问题解决了吗,paddle有对应的API了吗 torch.manual_seed

@GuoQuanhao
Copy link

paddle.framework.seed(seed) 好像在2.0版本找到了

@ToscanaGoGithub
Copy link

似乎是不能完全对齐的,就算是设置一样的随机数种子,根据这个种子初始化一些参数的时候仍然会与一些区别。只能依靠加载预训练模型来进行对齐

@paddle-bot-old
Copy link

Since you haven't replied for more than a year, we have closed this issue/pr.
If the problem is not solved or there is a follow-up one, please reopen it at any time and we will continue to follow up.
由于您超过一年未回复,我们将关闭这个issue/pr。
若问题未解决或有后续问题,请随时重新打开,我们会继续跟进。

@yyinsomnia
Copy link

paddle.seed(),用paddle.randint测试过自身是能能稳定随机。
但是用同样的seed,paddle和torch randint出来是不一样的,这个要对齐确实有点勉为其难了。

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

6 participants