GitLab

* add sdpa

* wip

* cleaning

* add ref

* yet more cleaning

* and more :)

* wip llama

* working llama

* add output_attentions=True support

* bigcode sdpa support

* fixes

* gpt-bigcode support, require torch>=2.1.1

* add falcon support

* fix conflicts falcon

* style

* fix attention_mask definition

* remove output_attentions from attnmaskconverter

* support whisper without removing any Copied from statement

* fix mbart default to eager renaming

* fix typo in falcon

* fix is_causal in SDPA

* check is_flash_attn_2_available in the models init as well in case the model is not initialized through from_pretrained

* add warnings when falling back on the manual implementation

* precise doc

* wip replace _flash_attn_enabled by config.attn_implementation

* fix typo

* add tests

* style

* add a copy.deepcopy on the config in from_pretrained, as we do not want to modify it inplace

* obey to config.attn_implementation if a config is passed in from_pretrained

* fix is_torch_sdpa_available when torch is not installed

* remove dead code

* Update src/transformers/modeling_attn_mask_utils.py
Co-authored-by: Arthur <48595927+ArthurZucker@users.noreply.github.com>

* Update src/transformers/modeling_attn_mask_utils.py
Co-authored-by: Arthur <48595927+ArthurZucker@users.noreply.github.com>

* Update src/transformers/modeling_attn_mask_utils.py
Co-authored-by: Arthur <48595927+ArthurZucker@users.noreply.github.com>

* Update src/transformers/modeling_attn_mask_utils.py
Co-authored-by: Arthur <48595927+ArthurZucker@users.noreply.github.com>

* Update src/transformers/modeling_attn_mask_utils.py
Co-authored-by: Arthur <48595927+ArthurZucker@users.noreply.github.com>

* Update src/transformers/models/bart/modeling_bart.py
Co-authored-by: Arthur <48595927+ArthurZucker@users.noreply.github.com>

* remove duplicate pretraining_tp code

* add dropout in llama

* precise comment on attn_mask

* add fmt: off for _unmask_unattended docstring

* precise num_masks comment

* nuke pretraining_tp in LlamaSDPAAttention following Arthur's suggestion

* cleanup modeling_utils

* backward compatibility

* fix style as requested

* style

* improve documentation

* test pass

* style

* add _unmask_unattended tests

* skip meaningless tests for idefics

* hard_check SDPA requirements when specifically requested

* standardize the use if XXX_ATTENTION_CLASSES

* fix SDPA bug with mem-efficient backend on CUDA when using fp32

* fix test

* rely on SDPA is_causal parameter to handle the causal mask in some cases

* fix FALCON_ATTENTION_CLASSES

* remove _flash_attn_2_enabled occurences

* fix test

* add OPT to the list of supported flash models

* improve test

* properly test on different SDPA backends, on different dtypes & properly handle separately the pad tokens in the test

* remove remaining _flash_attn_2_enabled occurence

* Update src/transformers/modeling_utils.py
Co-authored-by: Arthur <48595927+ArthurZucker@users.noreply.github.com>

* Update src/transformers/modeling_utils.py
Co-authored-by: Arthur <48595927+ArthurZucker@users.noreply.github.com>

* Update src/transformers/modeling_utils.py
Co-authored-by: Arthur <48595927+ArthurZucker@users.noreply.github.com>

* Update src/transformers/modeling_attn_mask_utils.py
Co-authored-by: Arthur <48595927+ArthurZucker@users.noreply.github.com>

* Update docs/source/en/perf_infer_gpu_one.md
Co-authored-by: Arthur <48595927+ArthurZucker@users.noreply.github.com>

* remove use_attn_implementation

* fix docstring & slight bug

* make attn_implementation internal (_attn_implementation)

* typos

* fix tests

* deprecate use_flash_attention_2=True

* fix test

* add back llama that was removed by mistake

* fix tests

* remove _flash_attn_2_enabled occurences bis

* add check & test that passed attn_implementation is valid

* fix falcon torchscript export

* fix device of mask in tests

* add tip about torch.jit.trace and move bt doc below sdpa

* fix parameterized.expand order

* move tests from test_modeling_attn_mask_utils to test_modeling_utils as a relevant test class is already there

* update sdpaattention class with the new cache

* Update src/transformers/configuration_utils.py
Co-authored-by: Arthur <48595927+ArthurZucker@users.noreply.github.com>

* Update src/transformers/models/bark/modeling_bark.py

* address review comments

* WIP torch.jit.trace fix. left: test both eager & sdpa

* add test for torch.jit.trace for both eager/sdpa

* fix falcon with torch==2.0 that needs to use sdpa

* fix doc

* hopefully last fix

* fix key_value_length that has no default now in mask converter

* is it flacky?

* fix speculative decoding bug

* tests do pass

* fix following #27907

---------
Co-authored-by: Arthur <48595927+ArthurZucker@users.noreply.github.com>

* translate

* update

* update

---------
Co-authored-by: jiaqiw <wangjiaqi50@huawei.com>

* translate work

* update

* update

* update [[autodoc]]

* Update callback.md

---------
Co-authored-by: jiaqiw <wangjiaqi50@huawei.com>

* docs: replace torch.distributed.run by torchrun

 `transformers` now officially support pytorch >= 1.10.
 The entrypoint `torchrun`` is present from 1.10 onwards.
Signed-off-by: Peter Pan <Peter.Pan@daocloud.io>

* Update src/transformers/trainer.py

with @ArthurZucker's suggestion
Co-authored-by: Arthur <48595927+ArthurZucker@users.noreply.github.com>

---------
Signed-off-by: Peter Pan <Peter.Pan@daocloud.io>
Co-authored-by: Arthur <48595927+ArthurZucker@users.noreply.github.com>

* translate deepspeed.md

* update

fixed the broken links belogs to dataset library of transformers

* translate

* update

* update

* translate model.md to chinese

* apply review suggestion
Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>

---------
Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>

* translate

* translate

* update

* translate perrf_torch_compile.md

* translate tf_xla.md

* update

* update

* update

* translate performance.md

* tranlsate performance.md and big_models.md

* update translation

* update review

* translate model_sharing.md

* translate llm_tutorial.md to chiense

* update wrong translation

* update _torctree.yml

* update typos

* update

* translate the en tokenizer_summary.md to Chinese

* revise WordPiece

* add to source/zh/_toctree.yml

* fixed links with 404

* make style

* translate run_scripts.md to chinese

* translate run_scripts.md to chinese

* translate run_scripts.md to chinese

* translate autoclass_tutorial.md  to chinese

* translate update

* tranlsate peft.md to chinese

* translate peft.md to chinese

* fix missing link

* translate task_summary.md to chinese

* update translation

* update translation

* fix _toctree.yml

* docs(zh): translate tflite.md

* docs(zh): add space around links

* Update docs/source/zh/tflite.md
Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>

---------
Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>

* translate traning.md

* update _tocree.yml

* update _tocree.yml

* update _tocree.yml

* docs(zh): translate serialization.md

* docs(zh): add space around links

* update translation

* fix problems mentioned in reviews

* docs(zh): translate custom_models.md

* minor fix in customer_models
Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>

---------
Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>

docs(zh): translate create_a_model.md

* translate preprocessing.md to Chinese

* update files fixing problems mentioned in review

* update files fixing problems mentioned in review

---------
Co-authored-by: jiaqiw <wangjiaqi50@huawei.com>

translate multilingual into Chinese
Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>

* update translation of pipeline_tutorial and preprocessing(Version1.0)

* update translation of pipeline_tutorial and preprocessing(Version2.0)

* update translation docs

* update to fix problems mentioned in review

---------
Co-authored-by: jiaqiw <wangjiaqi50@huawei.com>

docs: translate fast_tokenizers into Chinese

* translate accelerate page

* Update docs/source/zh/accelerate.md
Co-authored-by: Arthur <48595927+ArthurZucker@users.noreply.github.com>

---------
Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>
Co-authored-by: Arthur <48595927+ArthurZucker@users.noreply.github.com>

* translate installation to zh

* fix translation typo

* fix some samll bugs in readme

* Update docs/README.md
Co-authored-by: amyeroberts <22614925+amyeroberts@users.noreply.github.com>

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

* fix_all_language_quicktour

* give up ! before bash command

---------
Co-authored-by: lishukan <lishukan@dxy.cn>

* Rename index.mdx to index.md

* With saved modifs

* Address review comment

* Treat all files

* .mdx -> .md

* Remove special char

* Update utils/tests_fetcher.py
Co-authored-by: Lysandre Debut <lysandre.debut@reseau.eseo.fr>

---------
Co-authored-by: Lysandre Debut <lysandre.debut@reseau.eseo.fr>

* initial commit

* review changes

* post model PR merge

* updating doc

* Add ConvNeXt V2 to transformers
* TF model is separated from the PR to fix issues

translate index to zh
Co-authored-by: bfss <bfss@bfss.com>

* zh quicktour(#20095)

* add zh to doc workflow

* remove untranslation from toctree
Co-authored-by: BeifangSusu <BeifangSusu@bfss.com>