Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

[Bug Fix] fix qa pipeline tensor to numpy #31585

Merged
merged 2 commits into from
Jul 11, 2024
Merged

[Bug Fix] fix qa pipeline tensor to numpy #31585

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
904a13d
fix qa pipeline
jiqing-feng Jun 25, 2024
40754c5
fix tensor to numpy
jiqing-feng Jun 25, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion src/transformers/pipelines/question_answering.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -118,7 +118,7 @@ def select_starts_ends(
max_answer_len (`int`): Maximum size of the answer to extract from the model's output.
"""
# Ensure padded tokens & question tokens cannot belong to the set of candidate answers.
undesired_tokens = np.abs(np.array(p_mask) - 1)
undesired_tokens = np.abs(p_mask.numpy() - 1)
Copy link
Collaborator

@amyeroberts amyeroberts Jun 25, 2024 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Does this still work if you run the pipeline in jax?

from transformers import pipeline
pipe = pipeline("question-answering", model="hf-internal-testing/tiny-random-bert", framework="flax")
question = "What's my name?"
context = "My Name is Sasha and I live in Lyon."

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

It will raise a value error.
ValueError: Pipeline cannot infer suitable model classes from hf-internal-testing/tiny-random-bert.

Copy link
Contributor Author

@jiqing-feng jiqing-feng Jun 26, 2024 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Besides, temsor.numpy() has been already used in other pipelines like ASR

Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

OK, yes, looking into it we seem so assume either tf or pt everywhere in the pipeline, so even though I think this would break things for jax tensors it's not something we need to take account of at the moment. Thanks for testing!


if attention_mask is not None:
undesired_tokens = undesired_tokens & attention_mask
Expand Down