riaz.somc/transformers
1
0
Fork 0
mirror of https://github.com/huggingface/transformers.git synced 2025-08-02 19:21:31 +06:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity
14,819 Commits 892 Branches 232 Tags 2.1 GiB
87ae2a4632
Commit Graph

14819 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
frankenliu
87ae2a4632
fix bug:divide by zero in _maybe_log_save_evaluate() (#28251) 
Co-authored-by: liujizhong1 <liujizhong1@xiaomi.com>
 2024-01-02 14:19:42 +00:00
hoshi-hiyouga
502a10a6f8
Fix trainer saving safetensors: metadata is None (#28219) 
* Update trainer.py

* format
 2024-01-02 12:58:29 +00:00
Dean Wyatte
cad9f5c6cc
Update docs around mixing hf scheduler with deepspeed optimizer (#28223) 
update docs around mixing hf scheduler with deepspeed optimizer
 2024-01-02 11:48:17 +00:00
Stas Bekman
3cefac1d97
small typo (#28229) 
Update modeling_utils.py
 2023-12-26 21:52:10 +01:00
Sourab Mangrulkar
3b7675b2b8
fix FA2 when using quantization (#28203) 2023-12-26 08:36:41 +05:30
Younes Belkada
fa21ead73d
[Awq] Enable the possibility to skip quantization for some target modules (#27950) 
* v1

* add docstring

* add tests

* add awq 0.1.8

* oops

* fix test
 2023-12-25 11:06:56 +01:00
Younes Belkada
29e7a1e183
[Llava] Fix llava index errors (#28032) 
* fix llava index errors

* forward contrib credits from original implementation and fix

* better fix

* final fixes and fix all tests

* fix

* fix nit

* fix tests

* add regression tests

---------

Co-authored-by: gullalc <gullalc@users.noreply.github.com>
 2023-12-22 17:47:38 +01:00
lin yudong
68fa1e855b
update the logger message with accordant weights_file_name (#28181) 
Co-authored-by: yudong.lin <yudong.lin@funplus.com>
 2023-12-22 15:05:10 +00:00
Anindyadeep
74d9d0cebb
Fixing visualization code for object detection to support both types of bounding box. (#27842) 
* fix: minor enhancement and fix in bounding box visualization example

The example that was trying to visualize the bounding box was not considering an edge case,
where the bounding box can be un-normalized. So using the same set of code, we can not get
results with a different dataset with un-normalized bounding box. This commit fixes that.

* run make clean

* add an additional note on the scenarios where the box viz code works

---------

Co-authored-by: Anindyadeep <anindya@pop-os.localdomain>
 2023-12-22 13:24:40 +00:00
Yoach Lacombe
5da3db3fd5
[Whisper] Fix word-level timestamps with bs>1 or num_beams>1 (#28114) 
* fix frames

* use smaller chunk length

* correct beam search + tentative stride

* fix whisper word timestamp in batch

* add test batch generation with return token timestamps

* Apply suggestions from code review

Co-authored-by: amyeroberts <22614925+amyeroberts@users.noreply.github.com>
Co-authored-by: Sanchit Gandhi <93869735+sanchit-gandhi@users.noreply.github.com>

* clean a test

* make style + correct typo

* write clearer comments

* explain test in comment

---------

Co-authored-by: sanchit-gandhi <sanchit@huggingface.co>
Co-authored-by: amyeroberts <22614925+amyeroberts@users.noreply.github.com>
Co-authored-by: Sanchit Gandhi <93869735+sanchit-gandhi@users.noreply.github.com>
 2023-12-22 12:43:11 +00:00
Yih-Dar
c4df7c1668
Drop feature_extractor_type when loading an image processor file (#28195) 
fix

Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
 2023-12-22 13:19:04 +01:00
Yih-Dar
bb3bd44739
Fix the check of models supporting FA/SDPA not run (#28202) 
* add check_support_list.py

* fix

* fix

---------

Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
 2023-12-22 12:56:11 +01:00
Michael Feil
e37ab52dff
Bug: training_args.py fix missing import with accelerate with version accelerate==0.20.1 (#28171) 
* fix-accelerate-version

* updated with exported ACCELERATE_MIN_VERSION,

* update string in ACCELERATE_MIN_VERSION
 2023-12-22 11:41:35 +00:00
NielsRogge
c9fb250a25
Add Swinv2 backbone (#27742) 
* First draft

* More improvements

* More improvements

* Make all tests pass

* Remove script

* Update image processor

* Address comments

* Use new gradient checkpointing method

* Convert checkpoints, add integration test

* Do not keep aspect ratio for now

* Set keep_aspect_ratio=False for beit, add integration test

* Remove print statement
 2023-12-22 11:12:56 +00:00
Nicholas Neo
1ef86c4f56
Fix: [SeamlessM4T - S2TT] Bug in batch loading of audio in torch.Tensor format in the SeamlessM4TFeatureExtractor class (#27914) 
* fixes: code fixes on is_batched condition to also check for batched audio data in torch.Tensor format instead of only just checking for batched audio data in np.ndarray format

* Update src/transformers/models/seamless_m4t/feature_extraction_seamless_m4t.py

Co-authored-by: Yoach Lacombe <52246514+ylacombe@users.noreply.github.com>

* refactor: code refactoring to remove torch framework dependency

* docs: updated docstring to add torch tensor compatibility

* test: add test cases to incorporate torch tensor inputs

* test: ran make fix-copies for code conformity

* test: refactor test to separate the test_call into test_call_numpy and test_call_torch

---------

Co-authored-by: Yoach Lacombe <52246514+ylacombe@users.noreply.github.com>
 2023-12-22 10:47:30 +00:00
Dean Wyatte
548a8f6119
Fix ONNX export for causal LM sequence classifiers by removing reverse indexing (#28144) 
* normalize reverse indexing for causal lm sequence classifiers

* normalize reverse indexing for causal lm sequence classifiers

* normalize reverse indexing for causal lm sequence classifiers

* use modulo instead

* unify modulo-based sequence lengths
 2023-12-22 10:33:44 +00:00
Yih-Dar
71f460578d
Update docs/source/en/perf_infer_gpu_one.md (#28198) 
fix

Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
 2023-12-22 10:40:22 +01:00
Younes Belkada
3a8769f6a9
[Docs] Add 4-bit serialization docs (#28182) 
* add 4-bit serialization docs

* up

* up
 2023-12-22 10:18:32 +01:00
amyeroberts
3657748b4d
Update YOLOS slow test values (#28187) 
Update test values
 2023-12-21 18:17:07 +00:00
amyeroberts
cd1350ce9b
Fix slow backbone tests - out_indices must match stage name ordering (#28186) 
Indices must match stage name ordering
 2023-12-21 18:16:50 +00:00
Matt
260b9d2179
Even more TF test fixes (#28146) 
* Fix vision text dual encoder

* Small cleanup for wav2vec2 (not fixed yet)

* Small fix for vision_encoder_decoder

* Fix SAM builds

* Update TFBertTokenizer test with modern exporting + tokenizer

* Fix DeBERTa

* Fix DeBERTav2

* Try RAG fix but it's impossible to test locally

* Actually fix RAG now that I got FAISS working somehow

* Fix Wav2Vec2, add sermon

* Fix Hubert
 2023-12-21 15:14:46 +00:00
Arthur
f9a98c476c
[Mixtral & Mistral] Add support for sdpa (#28133) 
* some nits

* update test

* add support d\sd[a

* remove some dummy inputs

* all good

* style

* nits

* fixes

* fix more copies

* nits

* styling

* fix

* Update src/transformers/models/mistral/modeling_mistral.py

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

* add a slow test just to be sure

* fixup

---------

Co-authored-by: Younes Belkada <49240599+younesbelkada@users.noreply.github.com>
 2023-12-21 12:38:22 +01:00
Sanchit Gandhi
814619f54f
[Whisper] Use torch for stft if available (#26119) 
* [Whisper] Use torch for stft if available

* update docstring

* mock patch decorator

* fit on one line
 2023-12-21 11:04:05 +00:00
Joao Gante
7e93ce40c5
Fix input_embeds docstring in encoder-decoder architectures (#28168) 2023-12-21 11:01:54 +00:00
Poedator
4f7806ef7e
[bnb] Let's make serialization of 4bit models possible (#26037) 
* updated bitsandbytes.py

* rm test_raise_* from test_4bit.py

* add test_4bit_serialization.py

* modeling_utils bulk edits

* bnb_ver 0.41.3 in integrations/bitsandbytes.py

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

* @slow reinstated

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

* bnb ver 0.41.3 in  src/transformers/modeling_utils.py

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

* rm bnb version todo in  integrations/bitsandbytes.py

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

* moved 4b serialization tests to test_4bit

* tests upd for opt

* to torch_device

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

* ruff fixes to tests

* rm redundant bnb version check in mod_utils

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

* restore _hf_peft_config_loaded  modeling_utils.py::2188

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

* restore _hf_peft_config_loaded  test in modeling_utils.py::2199

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

* fixed NOT getattr(self, "is_8bit_serializable")

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

* setting model.is_4bit_serializable

* rm separate fp16_statistics arg from set_module...

* rm else branch in integrations::bnb::set_module

* bnb 4bit dtype check

* upd comment on 4bit weights

* upd tests for FP4 safe

---------

Co-authored-by: Younes Belkada <49240599+younesbelkada@users.noreply.github.com>
 2023-12-21 11:54:44 +01:00
Dean Wyatte
e268d7e5dc
disable test_retain_grad_hidden_states_attentions on SeamlessM4TModelWithTextInputTest (#28169) 
disable retain_grad_hidden_states_attentions on SeamlessM4TModelWithTextInputTest
 2023-12-21 08:39:44 +01:00
amyeroberts
1d77735947
Fix yolos resizing (#27663) 
* Fix yolos resizing

* Update tests

* Add a test
 2023-12-20 20:55:51 +00:00
Joao Gante
45b70384a7
Generate: fix speculative decoding (#28166) 
Co-authored-by: Merve Noyan <merveenoyan@gmail.com>
 2023-12-20 18:55:35 +00:00
Steven Liu
01c081d138
[docs] Trainer docs (#28145) 
* fsdp, debugging, gpu selection

* fix hfoption

* fix
 2023-12-20 10:37:23 -08:00
amyeroberts
ee298a16a2
Align backbone stage selection with out_indices & out_features (#27606) 
* Iteratre over out_features instead of stage_names

* Update for all backbones

* Add tests

* Fix

* Align timm backbone behaviour with other backbones

* Fix tests

* Stricter checks on set out_features and out_indices

* Revert back stage selection logic

* Remove out-of-order logic

* Document restriction in docstrings
 2023-12-20 18:33:17 +00:00
amyeroberts
224ab70969
Update FA2 exception msg to point to hub discussions (#28161) 
* Update FA2 exception msg to point to hub discussions

* Use path for hub url
 2023-12-20 16:52:16 +00:00
Yih-Dar
9924df9eb2
Avoid unnecessary warnings when loading CLIPConfig (#28108) 
* fix

---------

Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
 2023-12-20 17:24:53 +01:00
Yih-Dar
7938c8c836
Fix weights not properly initialized due to shape mismatch (#28122) 
* fix

---------

Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
 2023-12-20 14:20:02 +01:00
peter-sk
769a9542de
move code to Trainer.evaluate to enable use of that function with multiple datasets (#27844) 
* move code to Trainer.evaluate to enable use of that function with multiple datasets

* test

* update doc string

* and a tip

* forgot the type

---------

Co-authored-by: Prof. Peter Schneider-Kamp <jps@ordbogen.com>
 2023-12-20 10:55:56 +01:00
Jong-hun Shin
cd9f9d63f1
[gpt-neox] Add attention_bias config to support model trained without attention biases (#28126) 
* add attention_bias hparam for a model trained without attention biases

* fix argument documentation error
 2023-12-20 10:05:32 +01:00
Sourab Mangrulkar
def581ef51
Fix FA2 integration (#28142) 
* fix fa2

* fix FA2 for popular models

* improve warning and add Younes as co-author

Co-Authored-By: Younes Belkada <49240599+younesbelkada@users.noreply.github.com>

* Update src/transformers/modeling_utils.py

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

* fix the warning

* Add Tip

* typo fix

* nit

---------

Co-authored-by: Younes Belkada <49240599+younesbelkada@users.noreply.github.com>
Co-authored-by: Arthur <48595927+ArthurZucker@users.noreply.github.com>
 2023-12-20 14:25:07 +05:30
Abolfazl Shahbazi
b134f6857e
Remove deprecated CPU dockerfiles (#28149) 
Signed-off-by: Abolfazl Shahbazi <abolfazl.shahbazi@intel.com>
 2023-12-20 05:51:35 +01:00
Aaron Jimenez
38611086d2
[docs] Fix mistral link in mixtral.md (#28143) 
Fix mistral link in mixtral.md
 2023-12-19 10:34:14 -08:00
Mike Zellinger
23f8e4db77
Update modeling_utils.py (#28127) 
In docstring for PreTrainedModel.resize_token_embeddings, correct definition of new_num_tokens parameter to read "the new number of tokens" (meaning the new size of the vocab) rather than "the number of new tokens" (number of newly added tokens only).
 2023-12-19 09:07:57 -08:00
Arthur
4a04b4ccca
[Mixtral] Fix loss + nits (#28115) 
* default config should not use sliding window

* update the doc

* nits

* add a proper test

* update

* update

* update expected value

* Update src/transformers/tokenization_utils_fast.py

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

* convert to float

* average then N**2

* comment

* revert nit

* good to fo

* fixup

* Update tests/models/mixtral/test_modeling_mixtral.py

Co-authored-by: Lysandre Debut <hi@lysand.re>

* revert unrelated change

---------

Co-authored-by: Younes Belkada <49240599+younesbelkada@users.noreply.github.com>
Co-authored-by: Lysandre Debut <hi@lysand.re>
 2023-12-19 17:31:54 +01:00
Joao Gante
ac974199c8
Generate: speculative decoding (#27979) 
* speculative decoding

* fix test

* space

* better comments

* remove redundant test

* test nit

* Apply suggestions from code review

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

* PR comments

---------

Co-authored-by: amyeroberts <22614925+amyeroberts@users.noreply.github.com>
 2023-12-19 13:58:30 +00:00
amyeroberts
bd7a356135
Update split string in doctest to reflect #28087 (#28135) 2023-12-19 13:55:09 +00:00
qihqi
5aec50ecaf
When save a model on TPU, make a copy to be moved to CPU (#27993) 
* When save a model, make a copy to be moved to CPU, dont move the original
model

* make deepcopy inside of _save_tpu

* Move to tpu without copy
 2023-12-19 10:08:51 +00:00
Aaron Jimenez
4edffda636
[Doc] Fix token link in What 🤗 Transformers can do (#28123) 
Fix token link
 2023-12-18 15:06:54 -08:00
Mike Salvatore
c52b515e94
Fix a typo in tokenizer documentation (#28118) 2023-12-18 19:44:35 +01:00
Steven Liu
a52e180a0f
[docs] General doc fixes (#28087) 
* doc fix friday

* deprecated objects

* update not_doctested

* update toctree
 2023-12-18 10:44:09 -08:00
Rockerz
08a6e7a702
Fix indentation error - semantic_segmentation.md (#28117) 
Update semantic_segmentation.md
 2023-12-18 12:47:54 -05:00
Matt
71d47f0ad4
More TF fixes (#28081) 
* More build_in_name_scope()

* Make sure we set the save spec now we don't do it with dummies anymore

* make fixup
 2023-12-18 15:26:03 +00:00
Lucain
0695b2421a
Remove warning if DISABLE_TELEMETRY is used (#28113) 
remove warning if DISABLE_TELEMETRY is used
 2023-12-18 16:18:01 +01:00
Daize Dong
7c5408dade
Disable jitter noise during evaluation in SwitchTransformers (#28077) 
* Disable jitter noise during evaluation

* Update outdated configuration information

* Formatting

* Add new line
 2023-12-18 15:08:55 +00:00