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

Huggingface models AlbertForMaskedLM and AlbertForQuestionAnswering got fail accuracy #1251

Open
mengfei25 opened this issue Jan 7, 2025 · 1 comment
Open

Huggingface models AlbertForMaskedLM and AlbertForQuestionAnswering got fail accuracy #1251

mengfei25 opened this issue Jan 7, 2025 · 1 comment
Labels
Accuracy amp_fp16 E2E huggingface training

Comments

@mengfei25
Copy link
Contributor

mengfei25 commented Jan 7, 2025 •
edited
Loading

🐛 Describe the bug

With rhel and suse container, the 2 models accuracy failed.

python benchmarks/dynamo/huggingface.py --accuracy --amp_fp16 -d xpu -n10 --training --only AlbertForMaskedLM --backend=inductor

Versions

Host OS: Ubuntu:22.04
Container OS: rhel 9.4 (rolling driver) and suse 15.6 (lts driver) with docker
python: 3.10
XPU_OPS: 9ed0a1a
TRITON_COMMIT_ID: e98b6fcb8df5b44eb0d0addb6767c573d37ba024
TORCH_COMMIT_ID: 4f8b7c4272db521f7ffc4070ce1bdece513d1183
TRANSFORMERS_VERSION: 243e186efbf7fb93328dd6b34927a4e8c8f24395
BUNDLE_VERSION: 2025.0.1.20241113
@mengfei25
Copy link
Contributor Author

Same issue on ARC WSL

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
Accuracy amp_fp16 E2E huggingface training
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
1 participant
@mengfei25