[Fix] Add PyTorch 2.6+ compatibility fixes #1654
Open
Conversation
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
This commit addresses compatibility issues with PyTorch 2.6+ and 2.7+ that cause runtime errors in MMEngine. **PyTorch 2.6+ JIT Compilation Fix:** - Add safe import mechanism for ZeroRedundancyOptimizer in zero_optimizer.py - Temporarily disable JIT compilation during distributed optimizer import - Apply fix for PyTorch >=2.6.0 where JIT compilation issues were introduced - Graceful fallback when distributed optimizers are unavailable - Resolves: RuntimeError during torch.distributed.optim import **PyTorch 2.6+ torch.load weights_only Fix:** - Add _safe_torch_load function in checkpoint.py with automatic version detection - Handle weights_only parameter changes with safe globals for numpy arrays - Fallback to weights_only=False for compatibility with existing checkpoints - Resolves: "Weights only load failed" errors when loading models **Key Features:** - Maintains full backward compatibility with older PyTorch versions - Automatic version detection and appropriate handling - Conservative approach: only applies fixes to versions that need them - Comprehensive error handling and user warnings - Follows MMEngine coding standards **Version Support:** - PyTorch 2.6+ JIT compilation issues handled - PyTorch 2.6+ weights_only parameter changes handled - Full compatibility maintained for PyTorch 1.6-2.5 **Files Changed:** - mmengine/optim/optimizer/zero_optimizer.py: Safe distributed optimizer import - mmengine/runner/checkpoint.py: Safe torch.load with weights_only handling
|
|
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
This commit addresses compatibility issues with PyTorch 2.6+ and 2.7+ that cause runtime errors in MMEngine.
PyTorch 2.6+ JIT Compilation Fix:
PyTorch 2.6+ torch.load weights_only Fix:
Key Features:
Version Support:
Files Changed:
Thanks for your contribution and we appreciate it a lot. The following instructions would make your pull request more healthy and more easily get feedback. If you do not understand some items, don't worry, just make the pull request and seek help from maintainers. By the way, if you're not familiar with how to use pre-commit to fix lint issues or add unit tests, please refer to Contributing to OpenMMLab.
Motivation
Please describe the motivation of this PR and the goal you want to achieve through this PR.
Modification
Please briefly describe what modification is made in this PR.
BC-breaking (Optional)
Does the modification introduce changes that break the backward-compatibility of the downstream repos?
If so, please describe how it breaks the compatibility and how the downstream projects should modify their code to keep compatibility with this PR.
Use cases (Optional)
If this PR introduces a new feature, it is better to list some use cases here, and update the documentation.
Checklist
NOTE:this pr is generated by claude code