FIX Use torch.long instead of torch.int in LoftQ for PyTorch versions <2.x #1320
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
Solves huggingface#1307 For PyTorch < v2.x, using torch.int does not work for indexing, thus using torch.long.
|
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. |
younesbelkada
approved these changes
Jan 8, 2024
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Outdated suggestions cannot be applied.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.
Resolves #1307
For PyTorch versions below 2, the usage of torch.int instead of torch.long when indexing inside of
dequantize_blockfor LoftQ leads to an error. The fix consists of using torch.long.I tested the change with my local GPU for both PyTorch 1.13 and 2.1 and the tests passed for both.
Note that initially, it seemed to be an issue with seq2seq models, until it was discovered to be PyTorch-related. This is why I added the seq2seq tests. Even though the tests not really necessary, I kept them in.