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

How to debug margin and scale? #23

Open
AITTSMD opened this issue Aug 15, 2018 · 3 comments
Open

How to debug margin and scale? #23

AITTSMD opened this issue Aug 15, 2018 · 3 comments

Comments

@AITTSMD
Copy link

AITTSMD commented Aug 15, 2018

Hi, thanks for your great job. I wonder how I can debug the value of margin and scale to get better result.I use the default setting(m=0.35, scale=30)on my face recognition dataset, the final training loss is about 3 and it can't decrease. So I come here to ask the quesion, thank you~

@happynear
Copy link
Owner

The final loss value in AM-Softmax is meaningless. It doesn't matter what the final loss is.

@AITTSMD
Copy link
Author

AITTSMD commented Aug 15, 2018

@happynear okay,thank you~Can you explain the meaning of scale?How to adjust the value of scale to get better result?

@shiyuanyin
Copy link

shiyuanyin commented Dec 6, 2018

@AITTSMD
您好,
请问我想用 arcface 损失函数是将您的caffe 版本中的 将这一层, name: "label_specific_margin"
type: "LabelSpecificAdd" type改动,LabelSpecificHardMargin, 不需要参数设置,

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

3 participants