transformers
Fix many HPU failures in the CI
#39066
Merged

Fix many HPU failures in the CI #39066

SunMarc merged 14 commits into main from fix-hpu-errors
IlyasMoutawwakil
IlyasMoutawwakil more torch.hpu patches
9a93570b
IlyasMoutawwakil increase top_k because it results in flaky behavior when Tempreture, …
732144d1
IlyasMoutawwakil remove temporal fix
d02275b8
IlyasMoutawwakil fix scatter operation when input and src are the same
dd7e0c79
IlyasMoutawwakil trigger
9aae39d7
HuggingFaceDocBuilderDev
IlyasMoutawwakil fix and reduce
d37ffc7e
IlyasMoutawwakil skip finding batch size as it makes the hpu go loco
8990c35b
IlyasMoutawwakil fix fsdp (yay all are passing)
0655de41
IlyasMoutawwakil fix checking equal nan values
f30b437f
IlyasMoutawwakil style
b74508bc
IlyasMoutawwakil remove models list
ed974653
IlyasMoutawwakil IlyasMoutawwakil marked this pull request as ready for review 210 days ago
IlyasMoutawwakil IlyasMoutawwakil requested a review from ydshieh ydshieh 210 days ago
ydshieh
ydshieh commented on 2025-06-27
ydshieh
ydshieh commented on 2025-06-27
IlyasMoutawwakil order
daaec7b7
IlyasMoutawwakil rename to cuda_extensions
2db94c02
ydshieh
ydshieh commented on 2025-06-30
ydshieh
ydshieh commented on 2025-06-30
ydshieh
ydshieh commented on 2025-06-30
ydshieh
ydshieh approved these changes on 2025-06-30
ArthurZucker
ArthurZucker approved these changes on 2025-07-01
IlyasMoutawwakil Update src/transformers/trainer.py
6ee666c8
IlyasMoutawwakil
IlyasMoutawwakil
SunMarc
SunMarc approved these changes on 2025-07-03
SunMarc SunMarc merged 18e0cae2 into main 204 days ago
SunMarc SunMarc deleted the fix-hpu-errors branch 204 days ago

Login to write a write a comment.

Login via GitHub

Assignees
No one assigned
Labels
Milestone