intel / intel-xpu-backend-for-triton

OpenAI Triton backend for Intel® GPUs
MIT License
137 stars 42 forks source link

[E2E] Add pinned `transformers` and `timm_models` to nightly wheels #2101

Closed vlad-penkin closed 1 month ago

YarShev commented 1 month ago

It would be great to have explanation in the issue description why we have to use pinned versions.

pbchekin commented 1 month ago

It would be great to have explanation in the issue description why we have to use pinned versions.

We need to use pinned versions of E2E dependencies, because PyTorch uses them. See

Since we build nightly wheel for torch{vision,text,audio} for some time it is reasonable to also build nightly wheels for timm and transformers. This will allow running E2E locally.