Openequivariance integration#372
Open
allaffa wants to merge 9 commits into
Open
Conversation
added 9 commits
October 4, 2025 18:28
- Created test_openequivariance_basic.py with 11 integration tests - Created test_openequivariance_numerical_equivalence.py with 8 equivalence tests - Tests verify OpenEquivariance and e3nn produce identical results - All test functions use pytest_ naming convention for HydraGNN compatibility - Tests gracefully handle OpenEquivariance availability/unavailability - Comprehensive coverage of tensor operations, MACE integration, and configuration"
- Reformatted 6 files according to HydraGNN's black style guidelines - Added comprehensive integration documentation (OPENEQUIVARIANCE_INTEGRATION.md) - All code now follows project formatting standards - Ready for production use"
- Add explicit enable_openequivariance parameter to MACEStack.__init__() - Fix TypeError: __init__() got unexpected keyword argument 'enable_openequivariance' - Change test dtype from float64 to float32 for e3nn compatibility - Resolve TorchScript tensordot dtype mismatch errors in CI tests - Maintain backward compatibility with default enable_openequivariance=False"
- Remove enable_openequivariance from explicit parameter list to avoid conflict - Extract parameter from kwargs using pop() instead of get() - Resolves: TypeError: MACEStack.__init__() got multiple values for argument 'enable_openequivariance' - Parameter was being passed both positionally through **kwargs and explicitly - Now properly handles the parameter through kwargs extraction only"
Root Cause Analysis: - __getattr__ methods in OptimizedTensorProduct and OptimizedLinear were interfering with attribute creation - During __init__, assignments like self.tensor_product = ... would trigger __getattr__ causing AttributeError - This prevented the underlying e3nn objects from being created properly Solution: - Removed problematic __getattr__ delegation methods that caused recursion issues - Added explicit weight_numel property to OptimizedTensorProduct with intelligent fallback - Simplified attribute access to avoid interference with object initialization Test Results: ✅ pytest_train_model[ci.json-MACE] - PASSED ✅ pytest_train_model[ci_multihead.json-MACE] - PASSED ✅ pytest_train_mace_model_lengths[MACE] - PASSED ✅ pytest_train_equivariant_model[MACE] - PASSED This resolves the core CI failures related to 'weight_numel' attribute errors."
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.
No description provided.