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.

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

Jump to bottom

Refactor Temperature Scaling in Distillation Loss #444

Open
wants to merge 5 commits into
base: main
Choose a base branch
from
Open

Refactor Temperature Scaling in Distillation Loss #444

wants to merge 5 commits into from

Conversation

austin362667
Copy link
Collaborator

Summary

Addressed the part of issue raised in #441

Moving the scale temperature outside the distillation_loss_fn is fine as well. Keep the loss_fn simpler, and the rest can be handled in the forward function beforehand. Thanks to the advice by @Tcc0403

Testing Done

  • Hardware Type:
  • run make test to ensure correctness
  • run make checkstyle to ensure code style
  • run make test-convergence to ensure convergence

@austin362667 austin362667 changed the title Refactor temperature calculation Refactor temperature scaling Dec 9, 2024
@austin362667 austin362667 changed the title Refactor temperature scaling Refactor Temperature Scaling in Distillation Loss Dec 9, 2024
@ByronHsu
Copy link
Collaborator

please resolve the conflict

ByronHsu
ByronHsu requested changes Dec 10, 2024
View reviewed changes
Copy link
Collaborator

@ByronHsu ByronHsu left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

address conflict

@austin362667
Refactor temperatue calculation
25822f1 
Signed-off-by: Austin Liu <austin362667@gmail.com>
@austin362667 austin362667 force-pushed the austin362667/refactor_distill_temperature branch from d66f4d1 to 25822f1 Compare December 10, 2024 01:28
@austin362667
Copy link
Collaborator Author

@ByronHsu Sorry for the late response. Conflict addressed. Thank you~

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@ByronHsu ByronHsu ByronHsu requested changes

@lancerts lancerts lancerts approved these changes

Requested changes must be addressed to merge this pull request.

Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
3 participants
@austin362667 @ByronHsu @lancerts