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

Adaptive Asymmetry #93

Open
Jessegator opened this issue Aug 18, 2022 · 0 comments
Open

Adaptive Asymmetry #93

Jessegator opened this issue Aug 18, 2022 · 0 comments

Comments

@Jessegator
Copy link

Hi! First of all thank you for this interesting work!

I'm trying to apply this loss function to my multi-label dataset and to choose a good set of hyper-parameters. I'm wondering how the code dynamically change the hyper-parameter gamma_neg mentioned in Section 2.7. According to Eq.(11), there's a parameter named lambda, I'm not sure what value should I set to this one. Could you please share the setup of this part?

In addition, my dataset is a highly imbalanced one, and there are a lot of samples with labels which are all negative, which means during each iteration, there may appear a situation where all labels are all 0s. Do you think this technique would work on my dataset?

Thank you a lot!

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

1 participant