Fake: copy over grad attribute#82593
Closed
eellison wants to merge 4 commits intogh/eellison/303/basefrom
Closed
Conversation
[ghstack-poisoned]
This was referenced Aug 1, 2022
Contributor
🔗 Helpful links
❌ 32 New FailuresAs of commit eeac921 (more details on the Dr. CI page): Expand to see more
🕵️ 32 new failures recognized by patternsThe following CI failures do not appear to be due to upstream breakages
|
ezyang
approved these changes
Aug 1, 2022
ezyang
reviewed
Aug 1, 2022
| if type(t) is torch.nn.Parameter: | ||
| out = torch.nn.Parameter(out, requires_grad=out.requires_grad) # type: ignore[assignment] | ||
| if safe_is_leaf(t) and t.grad is not None: | ||
| out.grad = self.from_real_tensor(fake_mode, t.grad) |
Contributor
There was a problem hiding this comment.
what if we have grad on a non-leaf tensor 🤔
Contributor
Author
There was a problem hiding this comment.
TBH, i'm not really sure...
Contributor
There was a problem hiding this comment.
I think I'd just drop the leaf check
>>> torch.zeros(2, requires_grad=True)
tensor([0., 0.], requires_grad=True)
>>> x = torch.zeros(2, requires_grad=True)
>>> y = x * 2
>>> y.grad = torch.zeros(2)
>>> y
tensor([0., 0.], grad_fn=<MulBackward0>)
>>> y.grad
tensor([0., 0.])
Copy over the grad attribute... let me any suggestions of other tests/opinfos/crossrefs I should be doing to make this more comprehensive. [ghstack-poisoned]
Copy over the grad attribute... let me any suggestions of other tests/opinfos/crossrefs I should be doing to make this more comprehensive. [ghstack-poisoned]
Copy over the grad attribute... let me any suggestions of other tests/opinfos/crossrefs I should be doing to make this more comprehensive. [ghstack-poisoned]
Contributor
Author
|
@pytorchbot merge |
Collaborator
|
@pytorchbot successfully started a merge job. Check the current status here |
Contributor
|
Hey @eellison. |
facebook-github-bot
pushed a commit
that referenced
this pull request
Aug 3, 2022
Summary: Copy over the grad attribute... let me any suggestions of other tests/opinfos/crossrefs I should be doing to make this more comprehensive. Pull Request resolved: #82593 Approved by: https://github.com/ezyang Test Plan: contbuild & OSS CI, see https://hud.pytorch.org/commit/pytorch/pytorch/09a6d0b5bb6554926a4712c823e2da17bb7bd5b3 Reviewed By: kit1980 Differential Revision: D38359458 Pulled By: eellison fbshipit-source-id: 4a5d2e6bf18eb69c205c80513be000ec77820167
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
4 participants
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.
Stack from ghstack (oldest at bottom):
Copy over the grad attribute... let me any suggestions of other tests/opinfos/crossrefs I should be doing to make this more comprehensive.