ci: pull CPU-only torch wheels on GPU-less runners#385
Merged
Conversation
Set UV_TORCH_BACKEND=cpu on the uv install steps in tests.yml and doc.yaml. GPU-less ubuntu-latest runners install torch (via braindecode) with uv and no backend pin, so uv resolves the CUDA builds and pulls ~5 GB of unused nvidia CUDA/cuDNN libraries that also bloat the uv cache. Scoped to CI, not pyproject, so local and GPU installs are unaffected.
Contributor
📚 Documentation Preview📦 Download Documentation Artifact
💡 To enable live previews, add a |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What
Set
UV_TORCH_BACKEND: cpuon theuv pip installsteps intests.ymlanddoc.yamlso CI installs CPU-only torch.Why
Both jobs install eegdash (
braindecode[hub]→torch) viauvon GPU-lessubuntu-latestrunners with no backend pin. uv resolves the CUDA builds and pulls ~5 GB of nvidia CUDA/cuDNN libraries that never run, paid on every install and again on every uv cache upload/restore.Test
Forcing the Linux platform uv resolves in CI:
Notes
Scoped to CI, not
pyproject.toml, so local and GPU installs are unaffected.UV_TORCH_BACKENDneedsuv >= 0.6.0;setup-uv@v5defaults to latest. YAML validated.Ports facebookresearch/neuroai#163 to eegdash.