Update Loss Functions to Accept Tensor num_items_in_batch #38029
Update Loss Functions to Accept Tensor num_items_in_batch
ada5e324
iamsernine
marked this pull request as ready for review 304 days ago
SunMarc
dismissed these changes
on 2025-05-12
SunMarc
dismissed their stale review
301 days ago
Fix device mismatch by moving num_items_in_batch to loss device in fi…
2b5c012f
fix the ruff check
0825472d
delete the unused if stat
21783176
Merge branch 'main' into main
a4946e7b
fix the type problem
e3d1ed2f
Merge branch 'main' of github.com:NEREUScode/transformers
6eb9939a
Assignees
No one assigned