-
Notifications
You must be signed in to change notification settings - Fork 25
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
Comments
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 |
I also have this confusion, and my loss is nan. Do you konw why? Thanks! |
1 similar comment
I also have this confusion, and my loss is nan. Do you konw why? Thanks! |
#22 maybe you can get some help |
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
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.
The text was updated successfully, but these errors were encountered: