SUMIN/HuggingFace_transformer
1
0
Fork 0
Code Issues Pull Requests Actions 7 Packages Projects Releases Wiki Activity
13,259 Commits 5 Branches 239 Tags
ea91c2adca842da3d2f87e094504fa7d66a7008a
Commit Graph

13259 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Sanchit Gandhi
ea91c2adca [AutoModel] Add AutoModelForTextEncoding (#24305) 
* [AutoModel] Add AutoModelForTextEncoding

* add mt5

* add other models

* add to docs

* fix tf imports

* add tf to docs / init

* up

* fix inits

* add to dummy objects
 2023-06-23 10:01:37 +01:00
Weiming Zhao
feb83521ec [llama] Fix comments in weights converter (#24436) 
Explain the reason to clone tensor
 2023-06-22 20:38:53 -04:00
Yih-Dar
2c977e4a90 Save site-packages as cache in CircleCI job (#24424) 
* fix

* fix

* Upgrade complete!

---------

Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
 2023-06-22 23:16:35 +02:00
Sylvain Gugger
2834c17ad2 Clarify batch size displayed when using DataParallel (#24430) 2023-06-22 14:46:20 -04:00
Alex Hall
b6295b26c5 Refactor hyperparameter search backends (#24384) 
* Refactor hyperparameter search backends

* Simpler refactoring without abstract base class

* black

* review comments:
specify name in class
use methods instead of callable class attributes
name constant better

* review comments: safer bool checking, log multiple available backends

* test ALL_HYPERPARAMETER_SEARCH_BACKENDS vs HPSearchBackend in unit test, not module. format with black.

* copyright
 2023-06-22 14:28:25 -04:00
Matt
a1c4b63076 TF CI fix for Segformer (#24426) 
Fix segformer so compilation can figure out the channel dim
 2023-06-22 15:49:13 +01:00
Josh
754f61ca05 Update RayTune doc link for Hyperparameter tuning (#24422) 
Update outdated hyperlink hpo_train.md 

Link to RayTune search space API docs was outdated - have provided correct new link for docs.

Co-authored-by: Joshua Samuel <66880119+Joshsamuel101@users.noreply.github.com>
 2023-06-22 10:38:01 -04:00
Yih-Dar
8f2ef52fb6 Fix save_cache version in config.yml (#24419) 
fix

Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
 2023-06-22 16:18:16 +02:00
Younes Belkada
3ce3385c47 Revert "Fix gradient checkpointing + fp16 autocast for most models" (#24420) 
Revert "Fix gradient checkpointing + fp16 autocast for most models (#24247)"

This reverts commit 285a48011d.
 2023-06-22 16:11:27 +02:00
Younes Belkada
ebb62e8880 [bnb] Fix bnb serialization issue with new release (#24416) 
* fix bnb issue

* fixup

* revert and do simple patching instead

* add more details
 2023-06-22 15:40:38 +02:00
Yih-Dar
652ece0710 Skip test_conditional_generation_pt_pix2struct in Past CI (torch < 1.11) (#24417) 
fix

Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
 2023-06-22 15:34:13 +02:00
Matt
22fe73c378 TF safetensors reduced mem usage (#24404) 
* Slight comment cleanup

* Reduce peak mem usage when loading TF-format safetensor weights

* Tweak the PyTorch loading code to support lazy loading from safetensors

* Pass safe_open objects to the PyTorch loading function

* Do GPU transposes for speed

* One more tweak to reduce peak usage further

* One-line hasattr

* Fix bug when there's a shape mismatch

* Rename state_dict in the loading code to be clearer

* Use TF format everywhere for consistency
 2023-06-22 14:06:16 +01:00
Sanchit Gandhi
7e03e46934 [ASR pipeline] Check for torchaudio (#23953) 
* [ASR pipeline] Check for torchaudio

* add pip instructions

Co-authored-by: Sylvain Gugger <sylvain.gugger@gmail.com>

---------

Co-authored-by: Sylvain Gugger <sylvain.gugger@gmail.com>
 2023-06-22 13:48:49 +01:00
Yih-Dar
6ce6d62b6f Explicit arguments in from_pretrained (#24306) 
* fix

* fix

* fix

* fix

* fix

* fix

* fix

* fix

* fix

* fix

* fix

* fix

* fix

* fix

* fix

* fix

* fix

* fix

* fix

---------

Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
 2023-06-21 19:24:11 +02:00
Zach Mueller
127e81c272 Remove redundant code from TrainingArgs (#24401) 
Remove redundant code
 2023-06-21 11:51:27 -04:00
Matthijs Hollemans
cd927a4736 add word-level timestamps to Whisper (#23205) 
* let's go!

* initial implementation of token-level timestamps

* only return a single timestamp per token

* remove token probabilities

* fix return type

* fix doc comment

* strip special tokens

* rename

* revert to not stripping special tokens

* only support models that have alignment_heads

* add integration test

* consistently name it token-level timestamps

* small DTW tweak

* initial support for ASR pipeline

* fix pipeline doc comments

* resolve token timestamps in pipeline with chunking

* change warning when no final timestamp is found

* return word-level timestamps

* fixup

* fix bug that skipped final word in each chunk

* fix failing unit tests

* merge punctuations into the words

* also return word tokens

* also return token indices

* add (failing) unit test for combine_tokens_into_words

* make combine_tokens_into_words private

* restore OpenAI's punctuation rules

* add pipeline tests

* make requested changes

* PR review changes

* fix failing pipeline test

* small stuff from PR

* only return words and their timestamps, not segments

* move alignment_heads into generation config

* forgot to set alignment_heads in pipeline tests

* tiny comment fix

* grr
 2023-06-21 17:48:21 +02:00
Yih-Dar
0f968ddaa3 Check auto mappings could be imported via from transformers (#24400) 
* fix

* fix

* fix

* fix

* fix

* fix

---------

Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
 2023-06-21 17:31:57 +02:00
Zach Mueller
1a6fb930fb Clean up dist import (#24402) 2023-06-21 11:19:42 -04:00
Younes Belkada
285a48011d Fix gradient checkpointing + fp16 autocast for most models (#24247) 
* fix gc bug

* continue PoC on OPT

* fixes

* 🤯

* fix tests

* remove pytest.mark

* fixup

* forward contrib credits from discussions

* forward contrib credits from discussions

* reverting changes on untouched files.

---------

Co-authored-by: zhaoqf123 <zhaoqf123@users.noreply.github.com>
Co-authored-by: 7eu7d7 <7eu7d7@users.noreply.github.com>
 2023-06-21 17:04:59 +02:00
Meghan Cowan
1815d1865e [Trainer] Fix optimizer step on PyTorch TPU (#24389) 
* update optimizer step for tpu

* add comment
 2023-06-21 07:24:41 -04:00
Bearnardd
4c6e429589 fix type annotation for debug arg (#24033) 
* fix type annotation for debug arg

* fix TypeErorr
 2023-06-21 11:42:21 +01:00
Yih-Dar
16c7b16a0a byebye Hub connection timeout - Recast (#24399) 
fix

Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
 2023-06-21 12:36:34 +02:00
Joao Gante
5f0801d174 Generate: add SequenceBiasLogitsProcessor (#24334) 2023-06-21 11:14:41 +01:00
Yih-Dar
45f71d793d Add ffmpeg for doc_test_job on CircleCI (#24397) 
fix

Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
 2023-06-21 11:12:38 +02:00
Steven Liu
ad78d9597b [docs] Fix NLLB-MoE links (#24388) 
fix broken links
 2023-06-20 17:34:20 -07:00
Sergii Dymchenko
cb8f675510 Update deprecated torch.ger (#24387) 2023-06-20 20:21:13 -04:00
Sylvain Gugger
eb849f6604 Migrate doc files to Markdown. (#24376) 
* 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>
 2023-06-20 18:07:47 -04:00
Patrick von Platen
b0513b013b [Wav2Vec2 - MMS] Correct directly loading adapters weights (#24335) 
* Correct direct lang loading

* correct more

* revert black

* Use tie weights instead=

* add tests

* add tests

* make style
 2023-06-20 19:39:52 +02:00
Arthur
e5c760d636 [GPTNeoX] Nit in config (#24349) 
* add raise value error for attention size

* nits to fix test_config

* style
 2023-06-20 19:19:19 +02:00
Arthur
c2882403c4 [Whisper Docs] Nits (#24367) 
* nits

* config doc did not match

* Apply suggestions from code review

Co-authored-by: Sanchit Gandhi <93869735+sanchit-gandhi@users.noreply.github.com>

---------

Co-authored-by: Sanchit Gandhi <93869735+sanchit-gandhi@users.noreply.github.com>
 2023-06-20 19:18:52 +02:00
Yih-Dar
83dc5762e7 Skip a tapas (tokenization) test in past CI (#24378) 
fix

Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
 2023-06-20 18:35:45 +02:00
Yih-Dar
297d769d0e Better test name and enable pipeline test for pix2struct (#24377) 
* best test name forever

* fix

* fix

---------

Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
 2023-06-20 18:29:30 +02:00
Aaron Pham
6950f70b38 style: add BitsAndBytesConfig __repr__ function (#24331) 
* style: add repr to BitsAndBytesConfig

Signed-off-by: Aaron <29749331+aarnphm@users.noreply.github.com>

* chore: update pattern for __repr__

implement diff dict for __repr__ of BitsAndBytesConfig

Signed-off-by: Aaron <29749331+aarnphm@users.noreply.github.com>

---------

Signed-off-by: Aaron <29749331+aarnphm@users.noreply.github.com>
 2023-06-20 12:26:08 -04:00
Arthur
7feba74400 [Tokenizer doc] Clarification about add_prefix_space (#24368) 
* nits

* more details

* fixup

* Apply suggestions from code review

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

---------

Co-authored-by: amyeroberts <22614925+amyeroberts@users.noreply.github.com>
 2023-06-20 18:22:00 +02:00
Yih-Dar
0527c1c0ea Add a check in ImageToTextPipeline._forward (#24373) 
* fix

* fix

* fix

* Update src/transformers/pipelines/image_to_text.py

Co-authored-by: Nicolas Patry <patry.nicolas@protonmail.com>

* fix

---------

Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
Co-authored-by: Nicolas Patry <patry.nicolas@protonmail.com>
 2023-06-20 18:07:34 +02:00
Sylvain Gugger
dc4449918d Rename test to be more accurate (#24374) 2023-06-20 11:54:55 -04:00
Sylvain Gugger
a6b4d1ad83 Remove print statement 2023-06-20 11:14:29 -04:00
Sanchit Gandhi
6c1344444a [Whisper] Make tests faster (#24105) 2023-06-20 16:01:56 +01:00
Sanchit Gandhi
f924df3c7e [modelcard] add audio classification to task list (#24363) 2023-06-20 14:01:17 +01:00
Yih-Dar
c23d131eab Update tiny models for pipeline testing. (#24364) 
* fix

---------

Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
 2023-06-20 14:43:10 +02:00
Matt
56efbf4301 TensorFlow CI fixes (#24360) 
* Fix saved_model_creation_extended

* Skip the BLIP model creation test for now

* Fix TF SAM test

* Fix longformer tests

* Fix Wav2Vec2

* Add a skip for XLNet

* make fixup

* make fix-copies

* Add comments
 2023-06-20 12:59:21 +01:00
Llohann Dallagnol Speranca
183f442ba8 Fix resuming PeftModel checkpoints in Trainer (#24274) 
* Fix resuming checkpoints for PeftModels

Fix an error occurred when resuming a PeftModel from a training checkpoint. That was caused since PeftModel.pre_trained saves only adapter-related data while _load_from_checkpoint was expecting a torch sved model. This PR fix this issue and allows the adapter checkpoint to be loaded.

Resolves: #24252

* fix last comment

* fix nits

---------

Co-authored-by: younesbelkada <younesbelkada@gmail.com>
 2023-06-20 13:57:08 +02:00
Matt
0875b2509a Allow passing kwargs through to TFBertTokenizer (#24324) 2023-06-20 12:49:06 +01:00
Denis Ismailaj
cfc838dd4d Respect explicitly set framework parameter in pipeline (#24322) 
* Respect framework parameter

* Move check to pipeline()

* Add check inside infer_framework_load_model again
 2023-06-20 11:43:52 +01:00
Quentin Gallouédec
c5454eba9e Fix the order in GPTNeo's docstring (#24358) 
* Fix arg sort in docstring

* further order fix

* make style
 2023-06-19 18:59:35 +01:00
Ritesh Ghorse
20273ee214 [Doc Fix] Fix model name path in the transformers doc for AutoClasses (#24329) 
fix model name path

Co-authored-by: Ritesh Ghorse <riteshghorse@Riteshs-Air.attlocal.net>
 2023-06-19 17:26:55 +01:00
Aaron Pham
c003c8cb52 docs: add BentoML to awesome-transformers (#24344) 
* docs: add BentoML to awesome-transformers

Signed-off-by: Aaron <29749331+aarnphm@users.noreply.github.com>

* chore: add the project to the bottom of the line

Signed-off-by: Aaron <29749331+aarnphm@users.noreply.github.com>

---------

Signed-off-by: Aaron <29749331+aarnphm@users.noreply.github.com>
 2023-06-19 12:17:30 -04:00
Gema Parreño
52c4276e44 Fix link to documentation in Install from Source (#24336) 
Update __init__.py

Fix link to documentation to install Transformers from source 
Probably the title changed at some point from 'Installing' to 'Install'
 2023-06-19 17:12:55 +01:00
amyeroberts
7e71eb2ef7 Fix ImageGPT doctest (#24353) 
Fix doctest
 2023-06-19 15:23:29 +01:00
Yih-Dar
a4de24f691 Make AutoFormer work with previous torch version (#24357) 
fix

Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
 2023-06-19 16:02:06 +02:00