GitLab

Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>

* 📝 first draft

* 🖍 apply feedback

* Remove unused attributes

* Add link to blog and add clarification about input size

* Improve readability of the code
Co-authored-by: Niels Rogge <nielsrogge@Nielss-MacBook-Pro.local>

* Added type hints for PyTorch T5 model

* removed a type hint

* ran make style

* Add typing hints for base model class

* Add typing hints for causal LM model class

* Add typing hints for double heads model class

* Add typing hints for sequence classification model class

* Add typing hints for Main Layer

* Run fixup

* added type hints for BART model

* make fixup, adding imports to copied files

* Adding some missing types to cookiecutter

* Adding some missing types to cookiecutter

* Adding some missing types to cookiecutter
Co-authored-by: matt <rocketknight1@gmail.com>

* Add type hint for pt transfo_xl model

* Add type hint for tf transfo_xl model

* Add argument "cache_dir" for transformers.onnx

* Reformate files that can't pass CI.

* added type hints to xlnet PT

* added type hints to xlnet TF

* added type hints to xlnet TF

As mentioned: https://github.com/huggingface/transformers/issues/16277

* Update training.mdx

Fixed Error Raised Due to Wrongly Accessing Training Sample

* Ran make style

* Revert to Old Commit

* Apply suggestions from code review
Co-authored-by: Suraj Patil <surajp815@gmail.com>

* Draft a guide with our code quirks for new models

* Apply suggestions from code review
Co-authored-by: Suraj Patil <surajp815@gmail.com>
Co-authored-by: Joao Gante <joao@huggingface.co>

* Apply suggestions from code review
Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>
Co-authored-by: Suraj Patil <surajp815@gmail.com>
Co-authored-by: Joao Gante <joao@huggingface.co>
Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>

Co-authored-by: dinesh-GDK <dinesh.gna111@gmail.com1>

* Apply framework changes

* Fix italics

* Fix nits

* correct syntax
Co-authored-by: Omar Espejel <espejelomar@Omars-MacBook-Air.local>

* up

* up

* up

* fix

* yeh

* ups

* Empty test commit

* correct quicktour

* correct

* correct

* up

* up

* uP

* uP

* up

* up

* uP

* up

* up

* up

* up

* up

* up

* up

* up

* up

* up

* Update src/transformers/models/van/modeling_van.py

* finish

* apply suggestions

* remove folder

* revert to daily testing

* Removed spurious substraction

* Fixed condition checking for attention type

* Fixed sew_d copy of DeBERTa v2 attention

* Removed unused `p2p` attention type from DebertaV2-class models

* Fixed docs style

* Make add-new-model-like work without all frameworks installed

* A few fixes

* Last default frameworks

Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>

* 🖍 fix links in guides

* 🖍 apply feedback

* Aggressive PT/TF equivalence test on PT side

* Ugly fix for `TFTapasForQuestionAnswering`

* apply review suggestions
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>

* Make test_equivalence_pt_to_flax more aggressive

* Make test_equivalence_flax_to_pt more aggressive

* don't use to_tuple

* clean-up

* fix missing test cases + testing on GPU

* fix conversion

* fix `ValueError: assignment destination is read-only`

* Add type checking

* commit to revert later

* Fix

* fix

* fix device

* better naming

* clean-up
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>

* value check for typical sampling

* value check for typical sampling

* change from float to int comparison
Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>
Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>

* add unpack_inputs decorator for ctrl

* replace "past" with "past_key_values"
Co-authored-by: Johannes Kolbe <johannes.kolbe@tech.better.team>

* update XLM with tf decorator

* move to top decorator

* set unpack_inputs as top decorator
Co-authored-by: Louis Owen <yellow@Louis-Owen.local>

* Override _pad in LEDTokenizer

* Override _pad in LEDTokenizerFast

* add Copied from

* calling the super method

* add comment about -1
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>

* Change assertion to warning when passing past_key_value to T5 encoder

* lint

* Attention mask is important in the case of batching...

* Improve the fix.

* Making the sentence different enough that they exhibit different
predictions.

* Update expected slices for pillow > 9

* Add expected slices depending on pillow version

* Add different slices depending on pillow version for other models
Co-authored-by: Niels Rogge <nielsrogge@Nielss-MacBook-Pro.local>

integrations: mlflow: skip start_run() if a run is already active and sanity check on enabling integration (#16131)

* integrations: mlflow: skip start_run() call if a run is already active

* integrations: typo fix
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>