[Qwen2Audio] handle input ids expansion during processing #35534
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. |
ArthurZucker
approved these changes
Jan 7, 2025
Comment on lines
298
to
339
| # test the error when incorrect number of audio tokens | ||
| inputs["input_ids"] = torch.tensor( | ||
| [ | ||
| [ | ||
| 151644, | ||
| 8948, | ||
| 198, | ||
| 2610, | ||
| 525, | ||
| 264, | ||
| 10950, | ||
| 17847, | ||
| 13, | ||
| 151645, | ||
| 198, | ||
| 151644, | ||
| 872, | ||
| 198, | ||
| 14755, | ||
| 220, | ||
| 16, | ||
| 25, | ||
| 220, | ||
| 151647, | ||
| ] | ||
| + [151646] * 200 | ||
| + [ | ||
| 151648, | ||
| 198, | ||
| 3838, | ||
| 594, | ||
| 429, | ||
| 5112, | ||
| 30, | ||
| 151645, | ||
| 198, | ||
| 151644, | ||
| 77091, | ||
| 198, | ||
| ] | ||
| ] | ||
| ) |
1 task
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Outdated suggestions cannot be applied.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 adds input_id expansion to the processor.
Before:
"<|audio_bos|><|AUDIO|><|audio_eos|>Generate the caption in English:"→inputs_embedstensorNow:
"<|audio_bos|><|AUDIO|><|audio_eos|>Generate the caption in English:"→"<|audio_bos|><|AUDIO|>...<|AUDIO|><|audio_eos|>Generate the caption in English:"with as many<|AUDIO|>tokens as the number of audio embedded vectors after encoding and projectioninputs_embeds, directly with the correct shapeThis approach allows to remove the unnecessary step of creating a new embedding tensor in the model forward. Moreover, it simplifies the code as correct padding is directly handled at the processing stage.