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

Standardize losses to use ignore_index #7

Closed
isaaccorley opened this issue Feb 22, 2024 · 1 comment
Closed

Standardize losses to use ignore_index #7

isaaccorley opened this issue Feb 22, 2024 · 1 comment
Assignees

Comments

@isaaccorley
Copy link
Owner

isaaccorley commented Feb 22, 2024

Loss functions seem to use one of classes or ignore_index or even both. We should support both options for all loss functions. This generates confusion when using multiple loss functions (see microsoft/torchgeo#1896) This involves updating the following:

  • JaccardLoss
  • FocalLoss
  • DiceLoss (this uses both classes and ignore_index
  • TverskyLoss (this is just a subclass of DiceLoss)
@isaaccorley
Copy link
Owner Author

Closed in #11

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