make big model inference compatible with torch.compile #2609
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
|
The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update. |
|
Hi, I've hit the exact same error. Any news on this? |
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Outdated suggestions cannot be applied.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.
What does this PR do ?
This PR makes big model inference compatible with
torch.compile. To do that, we need to fixsend_to_device.In pytorch 2.2.1,
hasattris not implemented for TupleVariable, so we get an error inis_namedtuplefunction. I tried a few methods to not usehasattrbut I didn't manage to make it work (e.g. try except block). If we bypass theis_namedtuplefunction, compile works fine. In pytorch nightly (future 2.4), this is implemented but we hit another error ( breaking changes ... ). So, I need to investigate more.Fixes #2592
Draft for now to not forget
EDIT: Feel free to try this PR. It should work for the latest torch. However, it doesn't support namedtuple.