Refresh uv.lock to match pyproject.toml - #11
Open
swapnilashtekar wants to merge 1 commit into
Open
swapnilashtekar wants to merge 1 commit into
swapnilashtekar wants to merge 1 commit into
Conversation
The committed lockfile predates the cuda12 extra's torch pins and had drifted out of sync with pyproject.toml. It recorded torch 2.10.0 while pyproject has declared torch==2.3.0 / torchvision==0.18.0 under the cuda12 extra since those pins were added. `uv lock --check` against the old file passed only because it never recorded the cuda12 requirements in the first place. Regenerating picks them up. Because extras are additive, a base install and the cuda12 extra have to agree on one torch version, so the pin propagates to everyone: torch 2.10.0 -> 2.3.0 torchvision 0.25.0 -> 0.18.0 triton 3.6.0 -> 2.3.0 plus the matching cu121 runtime packages. This is the version pyproject has been asking for all along; the pins exist so OpenMM's JIT toolchain stays on a CUDA <=12.2 build (see the comment in pyproject.toml about PTX error 222). The base range >=2.2,<2.11 permits it. Verified on torch 2.3.0+cu121: all proteogram.v2 modules import, the full dependency stack including OpenMM 8.4 loads, every torchvision weight enum used by Img2Vec resolves, torch.cuda.is_available() is True, and no torch API newer than 2.3 is used anywhere in proteogram/ or scripts/. Splitting this out so the version change is reviewable on its own rather than riding along in an unrelated feature branch.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
The committed lockfile predates the cuda12 extra's torch pins and had drifted out of sync with pyproject.toml. It recorded torch 2.10.0 while pyproject has declared torch==2.3.0 / torchvision==0.18.0 under the cuda12 extra since those pins were added.
uv lock --checkagainst the old file passed only because it never recorded the cuda12 requirements in the first place.Regenerating picks them up. Because extras are additive, a base install and the cuda12 extra have to agree on one torch version, so the pin propagates to everyone:
torch 2.10.0 -> 2.3.0
torchvision 0.25.0 -> 0.18.0
triton 3.6.0 -> 2.3.0
plus the matching cu121 runtime packages. This is the version pyproject has been asking for all along; the pins exist so OpenMM's JIT toolchain stays on a CUDA <=12.2 build (see the comment in pyproject.toml about PTX error 222). The base range >=2.2,<2.11 permits it.
Verified on torch 2.3.0+cu121: all proteogram.v2 modules import, the full dependency stack including OpenMM 8.4 loads, every torchvision weight enum used by Img2Vec resolves, torch.cuda.is_available() is True, and no torch API newer than 2.3 is used anywhere in proteogram/ or scripts/.
Splitting this out so the version change is reviewable on its own rather than riding along in an unrelated feature branch.