Whisper: move to tensor cpu before converting to np array at decode time #31954
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. |
amyeroberts
approved these changes
Jul 14, 2024
| token_ids = token_ids.numpy() | ||
| if hasattr(token_ids, "numpy"): | ||
| if "torch" in str(type(token_ids)): | ||
| token_ids = token_ids.cpu().numpy() |
There was a problem hiding this comment.
Following from this - will token_ids ever have a grad? In which case, this will also fail on the cpu call
There was a problem hiding this comment.
token_ids, the output of generate, will not have gradients :) generate is decorated with @no_grad
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Outdated suggestions cannot be applied.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?
Follow up to #27818
pytest --doctest-modules src/transformers/models/whisper/generation_whisper.py -vvstarted failing onmaindue to the PR above.In a nutshell, if Whisper was running on GPU, the generated tensors would also be on GPU. The new decoding code called
token_ids.numpy(), which failed if thetoken_idstensor was on GPU. This PR moves it to the CPU before the numpy conversion :)cc @sanchit-gandhi