XLAShardedTensor.to_local() support #9505
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
Hoomaaan
force-pushed
the
toLocal_wspec
branch
2 times, most recently
from
f0c89b9 to
933a964
Compare
bfolie
reviewed
Aug 2, 2025
Comment on lines
+22
to
+24
| @classmethod | ||
| def setUpClass(cls): | ||
| super().setUpClass() |
There was a problem hiding this comment.
I don't think this method is necessary if there's no additional setup logic
There was a problem hiding this comment.
I added this to maintain consistency with other tests in the spmd folder, but happy to remove it if you think simplicity is more important than consistency here.
bfolie
reviewed
Aug 2, 2025
| # All gradients should be 1.0 since we did a sum() | ||
| self.assertTrue(torch.allclose(local_tensor.grad, torch.ones_like(tensor))) | ||
|
|
||
| print("Gradient flow test successful") |
…ure proper XLA support and maintain consistency with PyTorch/XLA SPMD integration.
Hoomaaan
force-pushed
the
toLocal_wspec
branch
from
812a69a to
6dc2351
Compare
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Outdated suggestions cannot be applied.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.
The implementation adds a to_local() method to XLAShardedTensor class that converts a sharded tensor back to its local representation while preserving gradient information.
requires_grad property is preserved
gradients are properly calculated and maintained
backward pass works correctly through the local tensor
gradient values are accurately preserved