Don't save self in index backward
#25594
Closed
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
`self` isn't necessary for `index` backward, we only need the shape of `self`. Changing derivatives.yaml to use `zeros_like(self)` triggers a codepath in the codegen to only save the shape. Fixes pytorch#24853. Test Plan - I added a new test that is adapted from the code in pytorch#24853. I'm not sure what a more minimal example would look like because the bug is hard to trigger because of how autograd handles differential views.
pytorchbot
added
module: autograd
yf225
approved these changes
Sep 3, 2019
There was a problem hiding this comment.
@zou3519 has imported this pull request. If you are a Facebook employee, you can view this diff on Phabricator.
ezyang
approved these changes
Sep 3, 2019
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Outdated suggestions cannot be applied.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.
selfisn't necessary forindexbackward, we only need the shape ofself. Changing derivatives.yaml to usezeros_like(self)triggers acodepath in the codegen to only save the shape.
Fixes #24853.
Test Plan
Run time error with backward() after upgrading to pytorch 1.2.0 #24853. I'm not sure what a
more minimal example would look like because the bug is hard to trigger
because of how autograd handles differential views.