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

confusion concerning the kl loss #17

Open
HowToNameMe opened this issue Sep 27, 2019 · 5 comments
Open

confusion concerning the kl loss #17

HowToNameMe opened this issue Sep 27, 2019 · 5 comments

Comments

@HowToNameMe
Copy link

The output of fc1 can be negative while traning. And in KL loss computing procedure, negative value is not valid. So I wonder how will you explain or modify it.

@alike123
Copy link

我想问下,这个kl loss计算的对吗?不应该是他注释掉的那几行才是正确的吗?H(p)可以直接用了l1 loss代替吗

@JeanXL
Copy link

JeanXL commented Nov 10, 2019

I also have this confusion, and my loss is nan. Do you konw why? Thanks!

1 similar comment
@zhaoyuchinese
Copy link

I also have this confusion, and my loss is nan. Do you konw why? Thanks!

@vicwer
Copy link
Owner

vicwer commented Nov 25, 2019

#22 maybe you can get some help

@gengyanlei
Copy link

KL loss其实就包含了softmax cross entroy;

请问,作者您在训练前,对人脸裁剪对齐了么?

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