Hotfix: fix .diagonal() calls for keops kernel matrices.
#2590
Conversation
This file contains 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
Balandat
approved these changes
Sep 20, 2024
| # Test diagonal | ||
| d1 = kern1(x1, x1).diagonal(dim1=-1, dim2=-2) | ||
| d2 = kern2(x1, x1).diagonal(dim1=-1, dim2=-2) | ||
| self.assertLess(torch.norm(d1 - d2), 1e-4) |
There was a problem hiding this comment.
This only tests that what the diagonal calls returns are the same, not that they actually compute the diagonal. You should add sth like this here.
Suggested change
| self.assertLess(torch.norm(d1 - d2), 1e-4) | |
| self.assertLess(torch.norm(d1 - d2), 1e-4) | |
| self.assertTrue(torch.equal(k1.diag(), d1)) |
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Outdated suggestions cannot be applied.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.
[Fixes #2589]
Note to @jacobrgardner : as part of the 2.0 refactor where we remove
LazyEvaluatedKernelTensor, I want to add a method_symmetric_diagto all kernels that replaces thediag=Truemode of current kernels.When the kernel matrix isn't symmetric and we call diagonal (not sure if this ever happens in gpytorch, but technically we support it...) the best we can do is what we currently do in
KernelLinearOperator.diagonal.So this PR is more of a temporary hotfix that should cause any regressions in behavior, and I'll make sure diag is fast 2.0 with #2342