GitLab

* Fix TF example in quicktour

* Fix model.fit() and the dataset section too

* Fixed the revert by making sure that even the regexp can cover all
duplicates.

* Code simplification using hash.

* Fixing the `ident`.

* Fixing ignoring patterened duplicate names.

* Using `accelerate@find_tied_parameters` for from_pretrained

This is more correct there, since it handles meta device seemlessly
and we don't need to handle "non-duplicate" tensors (slices of each
other).

* Protecting accelerate.

* Update src/transformers/modeling_utils.py
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

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

docs: ko: `run_scripts` to Korean
Co-authored-by: Hyeonseo Yun <0525_hhgus@naver.com>
Co-authored-by: Gabriel Yang <gabrielwithhappy@gmail.com>
Co-authored-by: Sohyun Sim <96299403+sim-so@users.noreply.github.com>
Co-authored-by: Wonhyeong Seo <wonhseo@kakao.com>
Co-authored-by: Jungnerd <46880056+jungnerd@users.noreply.github.com>

* Test hf_hub 0.14.0rc1

* fix mocked tests

* package version

---------
Co-authored-by: Sylvain Gugger <Sylvain.gugger@gmail.com>
Co-authored-by: testbot <lucainp@hf.co>

fix transformers keys

* Raise error if `stride` is too high

* Clarify use of `stride`

* fix

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

Add an attribute to disable custom kernels in deformable detr in order to make the model ONNX exportable (#22918)

* add disable kernel option

* add comment

* fix copies

* add disable_custom_kernels to config

* Update src/transformers/models/deta/modeling_deta.py
Co-authored-by: amyeroberts <22614925+amyeroberts@users.noreply.github.com>

* Update src/transformers/models/deta/modeling_deta.py
Co-authored-by: amyeroberts <22614925+amyeroberts@users.noreply.github.com>

* Update src/transformers/models/deta/modeling_deta.py
Co-authored-by: amyeroberts <22614925+amyeroberts@users.noreply.github.com>

* style

* fix

---------
Co-authored-by: amyeroberts <22614925+amyeroberts@users.noreply.github.com>

docs: ko: tasks/summarization.mdx
Co-authored-by: Hyeonseo Yun <0525_hhgus@naver.com>
Co-authored-by: Jungnerd <46880056+jungnerd@users.noreply.github.com>
Co-authored-by: Wonhyeong Seo <wonhseo@kakao.com>
Co-authored-by: Nayeon Han <nayeon2.han@gmail.com>
Co-authored-by: Gabriel Yang <gabrielwithhappy@gmail.com>
Co-authored-by: Kihoon Son <75935546+kihoon71@users.noreply.github.com>

docs: ko: `tasks/masked_language_modeling.mdx` to Korean
Co-authored-by: Hyeonseo Yun <0525_hhgus@naver.com>
Co-authored-by: Gabriel Yang <gabrielwithhappy@gmail.com>
Co-authored-by: Sohyun Sim <96299403+sim-so@users.noreply.github.com>
Co-authored-by: Wonhyeong Seo <wonhseo@kakao.com>
Co-authored-by: Jungnerd <46880056+jungnerd@users.noreply.github.com>

* run_check_tiny_models

* update summary

* update mixin

* update pipeline_model_mapping

* update pipeline_model_mapping

* Update for gpt_bigcode

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

clap nits

* docs: ko: init: accelerate.mdx

* docs: ko: translated: accelerate.mdx

* docs: ko: revised: natural expression accelerate.mdx
Co-Authored-By: Gabriel Yang <gabrielwithhappy@gmail.com>

* docs: ko: revised: natural expression2 accelerate.mdx
Co-authored-by: Sohyun Sim <96299403+sim-so@users.noreply.github.com>

---------
Co-authored-by: Gabriel Yang <gabrielwithhappy@gmail.com>
Co-authored-by: Sohyun Sim <96299403+sim-so@users.noreply.github.com>

Adds FocalNet by Microsoft to transformers

---------
Co-authored-by: Niels Rogge <nielsrogge@Nielss-MacBook-Pro.local>
Co-authored-by: alaradirik <alaradirik@gmail.com>

Feature to convert videomae huge and small finetuned on kinetics and ssv2 added to the videomae to pytorch converter (#22788)

* Feature to convert videomae huge finetuned kinetics and videomae small finetuned kinetics and ssv2 added to videomae to pytorch converter

* Reformat convert_videomae_to_pytorch using black

* Value exception added for the possible videomae model architectures

* patch

* add test

* move tests

* cover more cases (will fail nw update the code)

* style

* fix

* Update src/transformers/models/sam/image_processing_sam.py
Co-authored-by: amyeroberts <22614925+amyeroberts@users.noreply.github.com>

* Update src/transformers/models/sam/image_processing_sam.py
Co-authored-by: amyeroberts <22614925+amyeroberts@users.noreply.github.com>

* add better check

---------
Co-authored-by: Younes Belkada <49240599+younesbelkada@users.noreply.github.com>
Co-authored-by: amyeroberts <22614925+amyeroberts@users.noreply.github.com>
Co-authored-by: younesbelkada <younesbelkada@gmail.com>

* fix

* fix

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

* add test update and docs edits

* docs edit suggestion

ddp fixes for stable lm training

* patch test with values

* lower tol

* Hardcode GELU as the intermediate activation for ESM

* Sneak a quick fix to the weight tying in too

* Make the call to gelu explicit

* fix: half inference error

norm_factor is still torch.float32 after using model.half

So I changed it to register_buffer so I can change it to torch.float16 after using model.half

* fix: Added a variable "persistent=False"

* run make style

* expose

* style

* add dummy object

* amazed by the quality of transformers CI

* fix code not exportable

* fix

* Update src/transformers/models/sam/modeling_sam.py
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

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

* Seq2SeqTrainingArgs overriding to_dict for GenerationConfig json support

* seq2seqTrainingArgs to_dict calling super method before handling genconf

fix bug of CLAP: https://github.com/LAION-AI/CLAP/issues/62

Updates Swin MIM output class to match other masked image modeling outputs

* fix

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

* support gpt neox generate with inputs embeds

* Update src/transformers/models/gpt_neox/modeling_gpt_neox.py

great thx for the suggestion!
Co-authored-by: Joao Gante <joaofranciscocardosogante@gmail.com>

---------
Co-authored-by: Lei Li <tobiaslee@qq.com>
Co-authored-by: Joao Gante <joaofranciscocardosogante@gmail.com>

* integration tests were not being run

* add tests for short input waveform

* rewrite test for long input

* even more betterer

* my bad

* oh boy

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

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

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

This predicate will become available in tensorflow starting with version
2.14.
Co-authored-by: Russell Power <power@google.com>