Remove orphaned nvidia-*-cu11 pins after torch 2.13.0 upgrade - #189
Merged
Merged
Conversation
torch 2.13.0 uses the CUDA 13 (cu13) wheel stack, so the explicitly pinned CUDA 11 packages (nvidia-cublas-cu11, nvidia-cuda-nvrtc-cu11, nvidia-cuda-runtime-cu11, nvidia-cudnn-cu11) are no longer required by any dependency and were installed as dead weight. Remove them. Verified: full requirements resolve (185 packages) and a baseline DQL training run succeeds in the Docker image with torch 2.13.0.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
Follow-up cleanup after the torch 2.13.0 upgrade (#186). Removes four explicitly-pinned CUDA 11 packages that are no longer needed by any dependency.
Why
torch==2.13.0uses the CUDA 13 (cu13) wheel stack (nvidia-cublas,nvidia-cudnn-cu13, etc.). The followingcu11pins inrequirements.dev.txtwere left over from a much older torch and are now dead weight — pip/uv installs them as explicit top-level requirements even though nothing depends on them:nvidia-cublas-cu11==11.10.3.66nvidia-cuda-nvrtc-cu11==11.7.99nvidia-cuda-runtime-cu11==11.7.99nvidia-cudnn-cu11==8.5.0.96Removing them avoids installing ~1.5 GB of unused CUDA 11 libraries alongside the current CUDA 13 stack.
Verification
cu11packages remain).python -m cyberbattle.agents.baseline.run ... --chain_size=4), withtorch 2.13.0+cu130loaded and sensible converging reward curves — confirming the torch 2.13.0 upgrade works for training.