Commit Graph

17063 Commits

Author SHA1 Message Date
Cyril Vallez
17806d11ba
Improve modular converter (#33991)
* improve modular

* style

* Update modular_model_converter.py

* pretty print warning

* style

* Support to remove unused classes as part of added dependencies as well

* nits

* correct bug

* add example

* style

* Add documentation
2024-10-08 14:53:58 +02:00
Matt
fb360a6c7a
BatchFeature.to() supports non-tensor keys (#33918)
* Fix issue in oneformer preprocessing

* [run slow] oneformer

* [run_slow] oneformer

* Make the same fixes in DQA and object detection pipelines

* Fix BatchFeature.to() instead

* Revert pipeline-specific changes

* Add the same check in Pixtral's methods

* Add the same check in BatchEncoding

* make sure torch is imported
2024-10-08 13:43:32 +01:00
Matt
3b44d2f042
Image pipelines spec compliance (#33899)
* Update many similar visual pipelines

* Add input tests

* Add ImageToText as well

* Add output tests

* Add output tests

* Add output tests

* OutputElement -> Output

* Correctly test elements

* make fixup

* fix typo in the task list

* Fix VQA testing

* Add copyright to image_classification.py

* Revert changes to VQA pipeline because outputs have differences - will move to another PR

* make fixup

* Remove deprecation warnings
2024-10-08 13:34:28 +01:00
Yoni Gozlan
e2001c3413
Add auto model for image-text-to-text (#32472)
* Add Auto model for image-text-to-text

* Remove donut from processing auto, add chameleon ti image text to text models

* add qwen2_vl and llava_onevision

* add pixtral to auto model for image-text-to-text

* add mllama and idefics3

* remove models in IGNORE_NON_AUTO_CONFIGURED

* add AutoModelForImageTextToText to tests and doc
2024-10-08 14:26:43 +02:00
Raushan Turganbay
0dbc7090ba
Processors: don't default padding side (#33942)
* don't default padding side

* fix
2024-10-08 10:58:49 +02:00
Arthur
a3add29097
Add support for __all__ and potentilly deleting functions (#33859)
* Add support for __all__ and potentailly deleting functions

* updates

* update

* nits

* remove dummies

* fix warning

* fixup

* style

* update

* fixup

* skip copied from when # skip

* remove log

* bring dummies back

* fixup

* remove copied from

* fixup

* remove warnings from `make fix-copies`

* fix doc issues

* nits

* Better error message !

* add support for more flexible naming!

* style

* breaking style?

* fix super() renaming issues

* del not needed when you don't call super().__init__()

* style

* no more fmt on :)

* properly remove `self`

* fixup

* fix

* doc nits

* add some doc 🫡
2024-10-08 10:19:17 +02:00
Raushan Turganbay
bead0fa8dc
Cache: slight change in naming (#32421)
* squash

* codestyle

* Update src/transformers/cache_utils.py

Co-authored-by: Joao Gante <joaofranciscocardosogante@gmail.com>

* propagate changes to all cache classes

* + whisper

* fix tests

* more fixes

* add deprecation warning

* fix copies

* address comments

* fix mistral also

* these didn't have "copied from"

---------

Co-authored-by: Joao Gante <joaofranciscocardosogante@gmail.com>
2024-10-08 09:43:40 +02:00
Yijun Lee
d6ba1ac041
🌐 [i18n-KO] Translated gemma.md to Korean (#33936)
* docs: ko: gemma.md

* feat: nmt draft

* fix: manual edits
2024-10-07 15:59:14 -07:00
Jiwook Han
46f146a2b5
🌐 [i18n-KO] Translated vit.md to Korean (#33884)
* docs: ko: model_doc/vit.md

* feat: nmt draft

* fix: manual edits

* fix: manual edits

* Update docs/source/ko/model_doc/vit.md

Co-authored-by: Yijun Lee <119404328+yijun-lee@users.noreply.github.com>

* Update docs/source/ko/model_doc/vit.md

Co-authored-by: Chulhwa (Evan) Han <cjfghk5697@ajou.ac.kr>

---------

Co-authored-by: Yijun Lee <119404328+yijun-lee@users.noreply.github.com>
Co-authored-by: Chulhwa (Evan) Han <cjfghk5697@ajou.ac.kr>
2024-10-07 15:35:11 -07:00
Jiwook Han
1ecca92f03
🌐 [i18n-KO] Translated swin2sr.md to Korean (#33795)
* ko: doc: model_doc/swin2sr.md

* feat: nmt draft

* Update docs/source/ko/model_doc/swin2sr.md

Co-authored-by: Yijun Lee <119404328+yijun-lee@users.noreply.github.com>

---------

Co-authored-by: Yijun Lee <119404328+yijun-lee@users.noreply.github.com>
2024-10-07 15:34:56 -07:00
boyunJang
8258219c4c
🌐 [i18n-KO] Translated auto.md to Korean (#33590)
* docs: ko: model_doc/auto.md

* feat: nmt draft

* fix: manual edits

* fix: resolve suggestions

Co-authored-by: wony617 <49024958+Jwaminju@users.noreply.github.com>
Co-authored-by: YONGSANG <71686691+4N3MONE@users.noreply.github.com>

* fix: resolve suggestions

---------

Co-authored-by: wony617 <49024958+Jwaminju@users.noreply.github.com>
Co-authored-by: YONGSANG <71686691+4N3MONE@users.noreply.github.com>
2024-10-07 15:34:45 -07:00
Chaewon Song
253a9a9d6f
🌐 [i18n-KO] Translated logging.md to Korean (#33543)
* docs: ko: main_classes/logging.md

* feat: nmt-draft

* fix: update toctree.yml

* Update docs/source/ko/main_classes/logging.md

Co-authored-by: Sungmin Oh <fabxoe.kor@gmail.com>

* Update docs/source/ko/main_classes/logging.md

Co-authored-by: HyeokJun SHIN <96534680+jun048098@users.noreply.github.com>

* Apply suggestions from code review

Co-authored-by: HyeokJun SHIN <96534680+jun048098@users.noreply.github.com>
Co-authored-by: Sungmin Oh <fabxoe.kor@gmail.com>

* Apply suggestions from code review

Co-authored-by: Ahnjj_DEV <ahnjj.dev@gmail.com>

---------

Co-authored-by: Sungmin Oh <fabxoe.kor@gmail.com>
Co-authored-by: HyeokJun SHIN <96534680+jun048098@users.noreply.github.com>
Co-authored-by: Ahnjj_DEV <ahnjj.dev@gmail.com>
2024-10-07 15:34:34 -07:00
Yijun Lee
178d707b7e
🌐 [i18n-KO] Translated chameleon.md to Korean (#33799)
* docs: ko: chameleon.md

* feat: nmt draft

* fix: manual edits

* fix: resolve suggestions

Co-authored-by: Jiwook Han <33192762+mreraser@users.noreply.github.com>
Co-authored-by: Chulhwa (Evan) Han <cjfghk5697@ajou.ac.kr>

---------

Co-authored-by: Jiwook Han <33192762+mreraser@users.noreply.github.com>
Co-authored-by: Chulhwa (Evan) Han <cjfghk5697@ajou.ac.kr>
2024-10-07 15:06:13 -07:00
Yijun Lee
13432f8409
🌐 [i18n-KO] Translated trainer.md to Korean (#33797)
* docs: ko: trainer.md

* feat: nmt draft

* fix: manual edits

* fix: resolve suggestions

Co-authored-by: Jiwook Han <33192762+mreraser@users.noreply.github.com>
Co-authored-by: Chulhwa (Evan) Han <cjfghk5697@ajou.ac.kr>

---------

Co-authored-by: Jiwook Han <33192762+mreraser@users.noreply.github.com>
Co-authored-by: Chulhwa (Evan) Han <cjfghk5697@ajou.ac.kr>
2024-10-07 15:05:57 -07:00
Yijun Lee
e9fbe62965
🌐 [i18n-KO] Translated pipelines_utils.md to Korean (#33809)
* docs: ko: pipelines_utils.md

* feat: nmt draft

* fix: manual edits
2024-10-07 15:05:17 -07:00
Yijun Lee
9c61ba2f25
🌐 [i18n-KO] Translated time_series_utils.md to Korean (#33806)
* docs: ko: time_series_utils.md

* feat: nmt draft

* fix: manual edits
2024-10-07 15:05:00 -07:00
Yijun Lee
9c8bd3fc1b
🌐 [i18n-KO] Translated esm.md to Korean (#33796)
* docs: ko: esm.md

* feat: nmt draft

* fix: manual edits
2024-10-07 13:39:22 -07:00
Yijun Lee
6996f2186a
🌐 [i18n-KO] Translated audio_utils.md to Korean (#33802)
* docs: ko: audio_utils.md

* feat: nmt draft

* fix: manual edits
2024-10-07 13:39:10 -07:00
Jiwook Han
410c73af1d
🌐 [i18n-KO] Translated swinv2.md to Korean (#33566)
* docs: ko: model_doc/swinv2.md

* feat: nmt draft

* fix: manual edits

* fix: manual edits
2024-10-07 12:50:43 -07:00
Yijun Lee
6c18cefed0
🌐 [i18n-KO] Translated gguf.md to Korean (#33764)
* docs: ko: gguf.md

* feat nmt draft

* fix: manual edits

* fix: resolve suggestions

Co-authored-by: Jiwook Han <33192762+mreraser@users.noreply.github.com>
Co-authored-by: Chulhwa (Evan) Han <cjfghk5697@ajou.ac.kr>

---------

Co-authored-by: Jiwook Han <33192762+mreraser@users.noreply.github.com>
Co-authored-by: Chulhwa (Evan) Han <cjfghk5697@ajou.ac.kr>
2024-10-07 12:49:08 -07:00
Michael Goin
c91fe85b78
Fix undefined default_config in configuration_utils.py (#33934) 2024-10-07 18:32:20 +02:00
Arthur
736c7cde51
[pytes collection] Fix flax test collection (#34004)
bit weird but to filter I had to use this
2024-10-07 18:11:13 +02:00
roy
55be7c4c48
Enable customized optimizer for DeepSpeed (#32049)
* transformers: enable custom optimizer for DeepSpeed

* transformers: modify error message

---------

Co-authored-by: datakim1201 <roy.kim@maum.ai>
2024-10-07 15:36:54 +02:00
Arthur
7bae833728
properly fix and RUN_SLOW (#33965)
* properly fix and RUN_SLOW

* lots of models were affected

* fix-copies

* more fixes
2024-10-07 14:45:57 +02:00
Kaito
e782e95e34
Fix Tensor + Embedding error in some cases when using SiglipVisionModel (#33994)
Fix Tensor + Embedding error in some cases

Co-authored-by: kaitolucifer <kaito.o@ghelia.com>
2024-10-07 11:17:34 +02:00
Arthur
9b4b0c07db
[Red CIs] Fix hub failures (#34001)
maybe setup should work?
2024-10-07 10:56:24 +02:00
Magnus
ad1a250719
[Docs] Add Developer Guide: How to Hack Any Transformers Model (#33979)
* docs: add example for separating q, k, v projections in SAM

* docs: How to Hack Any Transformers Model

* docs: remove changes from sam model docs

* Apply suggestions from code review

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

---------

Co-authored-by: Arthur <48595927+ArthurZucker@users.noreply.github.com>
2024-10-07 10:08:20 +02:00
NielsRogge
f5aeb7c1a5
[Docs] Improve VLM docs (#33393)
* Improve docs

* Update docs/source/en/model_doc/llava.md

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

* Update docs/source/en/model_doc/llava.md

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

* Address comment

* Address comment

* Improve pixtral docs

---------

Co-authored-by: amyeroberts <22614925+amyeroberts@users.noreply.github.com>
2024-10-07 09:54:07 +02:00
Cyril Vallez
1f33023cfa
Flash-attn performance: remove cuda sync during inference (#33570)
Switch conditions to use short-circuit during inference
2024-10-07 09:52:19 +02:00
Avishai Elmakies
4953ddf036
Add position ids in forward pass to opt model (#33121)
* start working on adding position ids

* add docs

* Refactor modeling_biogpt.py and modeling_opt.py for code consistency

* fix 2 PR comments

* move position_ids to end of args

* remove trailing white space

* add comment with TODO

* bug fix gradient checkpointing

* fixup

* missed on position_ids

* remove _attention_to_position_ids and refactor embedding class

* remove redundent code

---------

Co-authored-by: Avishai Elmakies <avishai.elma@cs.huji.ac.il>
2024-10-07 09:20:49 +02:00
TomLim
1bd604d11c
[WIP] Add Tokenizer for MyT5 Model (#31286)
* Initial commit for MyT5 model

* custom implementation of MyT5 tokenizer, unused files deleted

* unittest for myt5 tokenizer

* upadate of import structure and style

* removed remmanents of MyT5Config

* fixed docstrings

* Updates after review: filled documentaion file, new docstrings and tests added

* Fixed code style issues

* fixed copied from to refer to function

* updated loading myt5 tokenizer in tests, added sample byte map file to fixtures

* changes after review

* removed redundant copied from

* removed redundant copied from

* optimalization and loading model from hf

* [run_slow] myt5

* [run-slow] myt5

* Updated en documentation for myt5

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

---------

Co-authored-by: Arthur <48595927+ArthurZucker@users.noreply.github.com>
2024-10-06 10:33:16 +02:00
Anton Vlasjuk
5ef432e474
[TF] Fix Tensorflow XLA Generation on limited seq_len models (#33903)
* fix tf xla generation on limited seq_len models

* [run-slow] opt

* [run-slow] opt
2024-10-05 16:20:50 +02:00
Vladislav Bronzov
22e102ad98
Bug fix gguf qwen2moe (#33940)
* fix qwen2moe tensors mapping, add unit tests

* add expert tensor split logic, test refactoring

* small params refactoring

* add comment to tensor reshaping
2024-10-05 16:19:01 +02:00
Yehoshua Cohen
56be9f1925
add test for Jamba with new model jamba-tiny-dev (#33863)
* add test for jamba with new model

* ruff fix

---------

Co-authored-by: Yehoshua Cohen <yehoshuaco@ai21.com>
2024-10-05 16:03:12 +02:00
Adam Pocock
a7e4e1a77c
Updating char_to_token documentation to note behaviour when trim_offsets is True (#33919)
Updating char_to_token documentation.
2024-10-05 14:13:26 +02:00
Raushan Turganbay
612065efeb
Paligemma: fix static cache test (#33941)
* fix

* not flaky anymore + style
2024-10-05 09:47:37 +02:00
Joao Gante
38f9f10dd9
Cache: revert DynamicCache init for BC (#33861)
* tmp commit

* tmp commit

* make fixup

* missing removal

* fix condition

* fix end-to-end compilation

* if -> elif

* BC

* BC

* use @deprecate_kwarg("num_hidden_layers", version="4.47.0")

* wups the import

* 🥴

---------

Co-authored-by: Arthur Zucker <arthur.zucker@gmail.com>
2024-10-04 22:47:08 +02:00
Arthur
f92d354823
fix red check-copies (#33964) 2024-10-04 22:45:37 +02:00
pglorio
f319ba16fa
Add Zamba (#30950)
* Update index.md

* Rebase

* Rebase

* Updates from make fixup

* Update zamba.md

* Batched inference

* Update

* Fix tests

* Fix tests

* Fix tests

* Fix tests

* Update docs/source/en/model_doc/zamba.md

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

* Update docs/source/en/model_doc/zamba.md

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

* Update configuration_zamba.py

* Update src/transformers/models/zamba/modeling_zamba.py

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

* Update src/transformers/models/zamba/modeling_zamba.py

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

* Update src/transformers/models/zamba/modeling_zamba.py

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

* Update src/transformers/models/zamba/modeling_zamba.py

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

* Update modeling_zamba.py

* Update modeling_zamba.py

* Update modeling_zamba.py

* Update configuration_zamba.py

* Update modeling_zamba.py

* Update modeling_zamba.py

* Merge branch 'main' of https://github.com/Zyphra/transformers_zamba

* Update ZambaForCausalLM

* Update ZambaForCausalLM

* Describe diffs with original mamba layer

* Moved mamba init into `_init_weights`

* Update index.md

* Rebase

* Rebase

* Updates from make fixup

* Update zamba.md

* Batched inference

* Update

* Fix tests

* Fix tests

* Fix tests

* Fix tests

* Update docs/source/en/model_doc/zamba.md

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

* Update docs/source/en/model_doc/zamba.md

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

* Update configuration_zamba.py

* Update src/transformers/models/zamba/modeling_zamba.py

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

* Update src/transformers/models/zamba/modeling_zamba.py

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

* Update src/transformers/models/zamba/modeling_zamba.py

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

* Update src/transformers/models/zamba/modeling_zamba.py

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

* Update modeling_zamba.py

* Update modeling_zamba.py

* Update modeling_zamba.py

* Update configuration_zamba.py

* Update modeling_zamba.py

* Update modeling_zamba.py

* Merge branch 'main' of https://github.com/Zyphra/transformers_zamba

* Update ZambaForCausalLM

* Moved mamba init into `_init_weights`

* Update ZambaForCausalLM

* Describe diffs with original mamba layer

* make fixup fixes

* quality test fixes

* Fix Zamba model path

* circleci fixes

* circleci fixes

* circleci fixes

* circleci fixes

* circleci fixes

* circleci fixes

* circleci fixes

* circleci fixes

* circleci fixes

* Update

* circleci fixes

* fix zamba test from merge

* fix ValueError for disabling mamba kernels

* add HF copyright

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

* shared_transf --> shared_transformer

* Update src/transformers/models/zamba/modeling_zamba.py

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

* Update src/transformers/models/zamba/modeling_zamba.py

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

* Fixes

* Move attention head dim to config

* Fix circle/ci tests

* Update modeling_zamba.py

* apply GenerationMixin inheritance change from upstream

* apply import ordering

* update needed transformers version for zamba

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

* add contribution author

* add @slow to avoid CI

* Update src/transformers/models/zamba/modeling_zamba.py

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

* Define attention_hidden_size

* Added doc for attention_head_size

* trigger CI

* Fix doc of attention_hidden_size

* [run-slow] zamba

* Fixed shared layer logic, swapped up<->gate in mlp

* shared_transformer -> shared_transf

* reformat HybridLayer __init__

* fix docstrings in zamba config

* added definition of _get_input_ids_and_config

* fixed formatting of _get_input_ids_and_config

---------

Co-authored-by: root <root@node-4.us-southcentral1-a.compute.internal>
Co-authored-by: Arthur <48595927+ArthurZucker@users.noreply.github.com>
Co-authored-by: root <root@node-1.us-southcentral1-a.compute.internal>
Co-authored-by: Quentin Anthony <qganthony@yahoo.com>
2024-10-04 22:28:05 +02:00
Amit Garg
e3775539c8
PhiMoE (#33363)
* onboard phimoe model

* removed debug code

* added unit tests

* updated docs

* formatted

* fixed unit tests

* fixed test case

* fixed format

* refactored code

* fixed expected outputs in the integration tests

* Added a warning msg

* Addressed comments

* Addressed comments

* fixed test cases

* added paper link

* Addressed comments

* Refactored PhimoeForCausalLM forward fn

* Refactored PhimoeRotaryEmbedding class

* fixed test cases

* fixed testcase

* fixed test case

* Addressed comments

* fixed test cases

* fixed testcases

* Used cache position instead to get the seq len
2024-10-04 21:39:45 +02:00
Arthur
46579c0e77
hot fix self.position_embeddings->self.position_embedding (#33958) 2024-10-04 21:35:31 +02:00
Longjie Zheng
0d1692a49b
Fix attn mask ignore logic in training-time trace (#32613)
* fix attn mask logic for training-time trace

* add test

* fix

* fix

* fix

* fix

* fix

* format

* [run-slow] llama

* avoid accelearate

* [run-slow] llama
2024-10-04 19:00:45 +02:00
karan-uppal3
614660fdb9
Removed unnecessary transpose in Switch Transformer Routing (#33582)
removed switch transformer routing transpose
2024-10-04 17:39:03 +02:00
Mohamed Abu El-Nasr
78ef58325c
🔴 🚨 Resizing tokens embeddings: initialize from old embeddings' normal distribution. (#33325)
* intilize new embeddings from normal distrib

* Fix typo in comments

* Fix typo in comments

* Fix style

* Fix variables naming

* Add tests

* Fix style

* code consistency nit

* Add deepspeed support

* Add deepspeed support

* Conver embeddings weights to float32 before computations

* Add deepspeed tests

* Cover when vocab_size is smaller than embedding_size

* Style fix

* Add tests for vocab_size smaller than hiddin_size

* Style fix

* Nits in tests

* Nits in tests

* Check for deepspeed before importing it

* Increase vocab_size for positive definite covariance matrix test

* Add warning

* Add multivariate_resizing flag and implement resizing for lm_heads

* Fix typo

* Fix wrong bias indexing

* Fix bias is zero check

* remove multivariate_resizing flag from tests

* Intialize bias from old bias normal distribution

* Fixup

* Code usability

* Use mean_resizing instead of multivariate_resizing

* Fix up

* Fix comments and docs
2024-10-04 16:29:55 +02:00
jiqing-feng
b916efcb3c
Enables CPU AWQ model with IPEX version. (#33460)
* enable cpu awq ipex linear

* add doc for cpu awq with ipex kernel

* add tests for cpu awq

* fix code style

* fix doc and tests

* Update docs/source/en/quantization/awq.md

Co-authored-by: Marc Sun <57196510+SunMarc@users.noreply.github.com>

* Update tests/quantization/autoawq/test_awq.py

Co-authored-by: Marc Sun <57196510+SunMarc@users.noreply.github.com>

* fix comments

* fix log

* fix log

* fix style

---------

Co-authored-by: Marc Sun <57196510+SunMarc@users.noreply.github.com>
2024-10-04 16:25:10 +02:00
Matt
de4112e4d2
Add a section on writing tool templates to the chat template docs (#33924)
* Add a section on writing tool templates to the chat template docs

* Small cleanups
2024-10-04 14:40:44 +01:00
Arthur
2e719e35fd
[PR run-slow] (#33939)
* force latest torch

* Update .github/workflows/self-pr-slow-ci.yml

Co-authored-by: Yih-Dar <2521628+ydshieh@users.noreply.github.com>

---------

Co-authored-by: Yih-Dar <2521628+ydshieh@users.noreply.github.com>
2024-10-04 14:46:15 +02:00
Raushan Turganbay
061c2c4c38
Ignore keys on validate_rope (#33753)
* ignore keys on check rope

* add tests

* fix tests, so maybe better leave at logger lvl
2024-10-04 12:39:37 +02:00
Artyom Semidolin
4a173b88b5
[i18n-ru] Fixes typo in the README_ru.md (#33882) 2024-10-04 11:21:38 +02:00
Deepak Saldanha
b6a01df6e9
[Doc]: Broken link in Kubernetes doc (#33879)
* add relative path in .md and redirects to conf.py

* add redirects to conf.py and update .md

* modify links in .md
2024-10-04 11:20:56 +02:00