SIGN IN SIGN UP

๐Ÿค— Transformers: the model-definition framework for state-of-the-art machine learning models in text, vision, audio, and multimodal models, for both inference and training.

0 0 0 Python

๐Ÿšจ More V5 pipeline cleanup (#43325)

* Rename the feature extraction pipelines and remove question-answering

* make style

* Remove more refs to the question-answering pipelines

* Remove more refs to the question-answering pipelines

* More migration guide

* make fix-repo

* Correct the name imports in init.py

* Cleanup lots of refs to the pipelines

* make fix-repo

* Remove a bunch of doc refs too

* Remove from not_doctested.txt

* More cleanup of the QA pipelines

* More cleanup of the QA pipelines

* Catch more examples in the docs

* Catch more examples in the docs

* Remove refs to the argumenthandler too

* A few more refs to fill-mask

* Remove references in the pipeline model mappings

* Remove references in the pipeline model mappings

* Apply suggestions from code review

Co-authored-by: Anton Vlasjuk <[email protected]>

* FeatureExtractionPipelineTests -> EmbeddingPipelineTests

* Change run_task_tests() calls

* Restore fill_mask.py

* Restore fillmaskpipeline tests

* make fix-repo

* Revert a few more bits

* Restore FillMaskPipeline in pipelines/__init__

* Revert more fill-mask changes

* Revert pipeline names

* Revert filenames

* Revert filenames

* Revert filenames

* image-embedding -> image-feature-extraction

* more fixup

* more fixup

* Last few reverts for feature-extraction / image-feature-extraction

* Clear out image-to-image entirely

* Clear out any references to image-to-text

* Remove references to pipelines in Q&A and VQ&A tasks

* Fix PEFT task list too

* Cleanup more pipeline mappings

* One more to drop from the doccheck list

* Fix up the toctrees

---------

Co-authored-by: Anton Vlasjuk <[email protected]>
M
Matt committed
0133a75cd28b3a481e3d4b94fc5cb89f01fbbefb
Parent: 0ff46c9
Committed by GitHub <[email protected]> on 2/24/2026, 5:22:39 PM