Skip to content

Reduced focal loss implementation is incorrect for binary case #102

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.

Sign up for GitHub

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

Merged
merged 2 commits into from
Oct 11, 2024
Merged

Reduced focal loss implementation is incorrect for binary case #102

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
2c78305
Update functional.py
MrPrajwalB Oct 9, 2024
654e1fd
Update functional.py
MrPrajwalB Oct 9, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion pytorch_toolbelt/losses/functional.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -70,7 +70,7 @@ def focal_loss_with_logits(
if reduced_threshold is None:
focal_term = (1.0 - pt).pow(gamma)
else:
focal_term = ((1.0 - pt) / reduced_threshold).pow(gamma)
focal_term = ((1.0 - pt) / (1 - reduced_threshold)).pow(gamma) #the focal term continuity breaks when reduced_threshold not equal to 0.5. At pt equal to reduced_threshold, the value of piecewise function of focal term should be 1 from both sides .
focal_term = torch.masked_fill(focal_term, pt < reduced_threshold, 1)

loss = focal_term * ce_loss
Expand Down
Loading