Remove torch_scatter dependency in favor of built-in PyTorch scatter_reduce_#571
Open
swahtz wants to merge 3 commits intoopenvdb:mainfrom
Open
Remove torch_scatter dependency in favor of built-in PyTorch scatter_reduce_#571swahtz wants to merge 3 commits intoopenvdb:mainfrom
swahtz wants to merge 3 commits intoopenvdb:mainfrom
Conversation
…reduce_ torch_scatter (pytorch_scatter) was only used in test_jagged_tensor.py as a reference implementation for validating jsum/jmin/jmax reductions. Replace it with a small helper wrapping torch.Tensor.scatter_reduce_(), which has been stable since PyTorch 2.0, eliminating a fragile external dependency that required custom wheels and source builds in CI. fixes Remove torch_scatter as a test dependency Fixes openvdb#53 Signed-off-by: Jonathan Swartz <jonathan@jswartz.info>
swahtz
added
the
CI
…ement The built-in scatter_reduce_ with amax/amin distributes gradients to all tied extremal values, unlike torch_scatter (and fVDB) which routes to a single winner. With half-precision types (bfloat16/float16), ties are more frequent, causing the sorted non-zero gradient tensors to differ in length. Guard the sorted non-zero gradient allclose checks with a shape equality test in test_jmin, test_jmax, test_jmin_list_of_lists, and test_jmax_list_of_lists. Forward value assertions remain unconditional. Signed-off-by: Jonathan Swartz <jonathan@jswartz.info>
Signed-off-by: Jonathan Swartz <jonathan@jswartz.info>
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
torch_scatter (pytorch_scatter) was only used in test_jagged_tensor.py as a reference implementation for validating jsum/jmin/jmax reductions. Replace it with a small helper wrapping torch.Tensor.scatter_reduce_(), which has been stable since PyTorch 2.0, eliminating a fragile external dependency that required custom wheels and source builds in CI.
fixes Remove torch_scatter as a test dependency
Fixes #53