[Feature] Contiguous stacking of matching specs #960
Merged
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
facebook-github-bot
added
the
CLA Signed
tcbegley
commented
Mar 9, 2023
vmoens
approved these changes
Mar 9, 2023
|
We have some failing tests due to the device changes |
4 tasks
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Outdated suggestions cannot be applied.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.
Description
This PR changes the behaviour of
torch.stackwhen the arguments areTensorSpecorCompositeSpecvariants. It checks for equality among the specs being stacked, and if they all match, it actually performs the stack (rather than returning a lazy stacked spec).To facilitate this, I have added
squeezeandunsqueezemethods to all specs. Stacking is done by computing the new shape and then calling something likespec.clone().unsqueeze(stack_dim).expand(new_shape). I addedsqueezesince you can't haveunsqueezewithoutsqueeze😄Some of the tests probably need rethinking. I've made small changes to get them to pass, but there is new behaviour here which ideally would be tested for explicitly.