dnrops
/
transformers
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
11,235 Commits 481 Branches 161 Tags 1.2 GiB
Commit Graph

11235 Commits

Author SHA1 Message Date
Joao Gante d39f794eda
Generate: contrastive search cosmetic tweaks (#19871) 2022-10-25 14:43:06 +01:00
Davi Alves 0a77249178
Added translation of serialization.mdx to Portuguese Issue #16824 (#19869) 
* [ custom_models.mdx ] - Translated to Portuguese the custom models tutorial.

* [ run_scripts.mdx ] - Translated to Portuguese the run scripts tutorial.

* [ converting_tensorflow_models.mdx ] - Translated to Portuguese the converting tensorflow models tutorial.

* [ converting_tensorflow_models.mdx ] - Translated to Portuguese the converting tensorflow models tutorial.

* [ serialization.mdx ] - Translated to Portuguese the serialization tutorial.
 2022-10-25 09:34:28 -04:00
Guillaume Klein ab108a0e31
Add missing lang tokens in M2M100Tokenizer.get_vocab (#18416) 2022-10-25 09:18:24 -04:00
Yih-Dar 0bd6d9340e
Fix doctest for `GenerationMixin.contrastive_search` (#19863) 
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
 2022-10-25 14:51:16 +02:00
Alberto Mario Ceballos-Arroyo 371337a95b
Spanish translation of multiple_choice.mdx, question_answering.mdx. (#19821) 
* Translated multiple_choice.mdx, question_answering.mdx. Added them to _toctree.yml

* Added translation for a missed line.

* Update _toctree.yml as per Omar's suggestions

* Update multiple_choice.mdx as per Omar's comments

* Updt question_answering.mdx as per Omar's comments
 2022-10-24 20:11:34 -04:00
Sylvain Gugger d4eb52d13d
Refactor conversion function (#19799) 
* Refactor conversion function

* Remove dupe line

* Fixes

* Fixes

* Use the right variable...

* Fix last test
 2022-10-24 13:48:40 -04:00
Steven Liu 9ecb13d63a
add small updates only (#19847) 2022-10-24 10:18:20 -07:00
Yih-Dar 072ed01c38
Fix doctest for `MarkupLM` (#19845) 
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
 2022-10-24 17:54:23 +02:00
Rak Alexey 1f7e40d04f
Improve check copies (#19829) 
* print first diff line intead of first code part line

* fix style
 2022-10-24 11:24:18 -04:00
Yih-Dar 8b2501b4b9
Update `LEDModelIntegrationTests` expected values (#19841) 
* Update expected values

* fix style

Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
 2022-10-24 16:05:26 +02:00
Dhruv Singal 5cbf1fa8ca
fixed typo in fp16 training section for perf_train_gpu_one (#19736) 2022-10-24 10:04:28 -04:00
Yih-Dar 8db92dbe26
Fix nightly CircleCI (#19837) 
* Fix nightly CircleCI

* update

Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
 2022-10-24 16:00:02 +02:00
Davi Alves 743995e0e6
Added translation of converting_tensorflow_models.mdx to Portuguese Issue #16824 (#19824) 
* [ custom_models.mdx ] - Translated to Portuguese the custom models tutorial.

* [ run_scripts.mdx ] - Translated to Portuguese the run scripts tutorial.

* [ converting_tensorflow_models.mdx ] - Translated to Portuguese the converting tensorflow models tutorial.

* [ converting_tensorflow_models.mdx ] - Translated to Portuguese the converting tensorflow models tutorial.
 2022-10-24 09:50:16 -04:00
Rak Alexey d3f4cef74d
fix image2test args forwarding (#19648) 
* fix image2test args forwarding

* fix issues

* Proposing the update to the PR.

* Fixup.

Co-authored-by: Nicolas Patry <patry.nicolas@protonmail.com>
 2022-10-24 09:49:24 -04:00
zhou fan 3b419cfc6f
fix broken links in testing.mdx (#19820) 2022-10-24 09:48:02 -04:00
Yih-Dar 7ccd6fc47c
Fix OOM in Config doctest (#19840) 
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
 2022-10-24 15:33:00 +02:00
Angus L'Herrou 18adc40d87
replace reference to Datasets in metrics deprecation with Evaluate (#19812) 2022-10-24 09:25:57 -04:00
regisss 0b59ecdefd
Display the number of trainable parameters when lauching a training (#19835) 2022-10-24 09:15:52 -04:00
Ayaka 536f338441
[Doctest] Add `configuration_nezha.py` (#19810) 
* [Doctest] Add `configuration_nezha.py`

* Revert line order
 2022-10-24 13:50:43 +02:00
Ayaka f58b211ed3
[Doctest] Add `configuration_electra.py` (#19807) 2022-10-24 12:34:43 +02:00
Ayaka c949188b9d
[Doctest] Add `configuration_poolformer.py` (#19808) 2022-10-24 12:33:46 +02:00
Ayaka 82df83a96b
[Doctest] Add `configuration_plbart.py` (#19809) 
Additionally, I updated the doctest format to be consistent with BERT.
 2022-10-24 12:32:55 +02:00
sha016 22502ebb85
[Doctest] `MaskFormerConfig` doctest (#19817) 2022-10-24 11:08:32 +02:00
ydshieh 6f8064da6b install GitPython 2022-10-24 09:54:15 +02:00
Joao Gante 674f750a57
Generate: minor docstring fix (#19801) 2022-10-23 10:46:47 +01:00
Davi Alves 74b3eb3dea
Added translation of run_scripts.mdx to Portuguese Issue #16824 (#19800) 
* [ custom_models.mdx ] - Translated to Portuguese the custom models tutorial.

* [ run_scripts.mdx ] - Translated to Portuguese the run scripts tutorial.
 2022-10-21 17:38:35 -04:00
Yih-Dar 3436842102
Run some TF Whisper tests in subprocesses to avoid GPU OOM (#19772) 
* Run some TF Whisper tests in subprocesses to avoid GPU OOM

Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
 2022-10-21 21:59:18 +02:00
Joao Gante e0b825a8d0
Generate: contrastive search test updates (#19787) 
* contrastive search test updates

* make fixup
 2022-10-21 19:10:08 +01:00
Sylvain Gugger c4a997cd85
Use None to detect if truncation was unset (#19794) 
* Use None to detect if truncation was unset

* Fix repo consistency
 2022-10-21 12:53:37 -04:00
Serhii Hamotskyi 2e5c6f5975
Fix error/typo in docstring of TokenClassificationPipeline (#19798) 2022-10-21 12:53:16 -04:00
Alara Dirik cca51aa151
Fix image segmentation pipeline errors, resolve backward compatibility issues (#19768) 
* Fix panoptic segmentation and pipeline
* Update ImageSegmentationPipeline tests and reenable test_small_model_pt
* Resolve backward compatibility issues
 2022-10-21 18:09:58 +03:00
Sylvain Gugger b58d4f70f6
Fix nightly test setup (#19792) 2022-10-21 10:26:30 -04:00
Yih-Dar 3a1aeea3c5
Fix CTRL `test_torchscrip_xxx` CI by updating `_create_and_check_torchscript` (#19786) 
* Run inputs before trace

* Run inputs before trace

Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
 2022-10-21 16:23:13 +02:00
Hao Wang 31565ff0fd
Add sentencepiece to BertJapaneseTokenizer (#19769) 
* support sentencepiece for bertjapanesetokenizer

* add test vocab file for sentencepiece, bertjapanesetokenizer

* make BasicTokenizer be identical to transformers.models.bert.tokenization_bert.BasicTokenizer

* fix missing of \n in comment

* fix init argument missing in tests

* make spm_file be optional, exclude spiece.model from tests/fixtures, and add description comments

* make comment length less than 119

* apply doc style check
 2022-10-21 10:04:49 -04:00
Davi Alves 2ebf4e6a7b
[ custom_models.mdx ] - Translated to Portuguese the custom models tutorial. (#19779) 2022-10-21 09:48:19 -04:00
ftorres16 c1f009ad9a
Update training.mdx (#19791) 2022-10-21 09:46:44 -04:00
Sylvain Gugger 9151e649a5
Make public versions of private tensor utils (#19775) 
* Make public versions of private utils

* I need sleep
 2022-10-21 09:34:01 -04:00
Yih-Dar 3aaabaa214
Update `ImageToTextPipelineTests.test_small_model_tf` (#19785) 
* update expected values for the correct TF checkpoint

* Run test

* Clean up

* fix

Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
 2022-10-21 14:35:20 +02:00
Kashif Rasul 7487829a23
Added support for multivariate independent emission heads (#19453) 
* Added support for multivariate independent emission heads

* fix typo

* rename distr_cls

* scale is a vector for multivariate

* set affine transform event_dim

* fix typo

* added variable

* added beta in the config

* set beta

* remove beta-nll option in nll
 2022-10-21 08:32:10 -04:00
Matt a5da6f1817
Add warning about restarting runtime to import errors (#19774) 
* Add warning about restarting runtime to import errors

* Fix some linebreaks
 2022-10-21 11:52:29 +01:00
Yih-Dar 84f6bee5da
PT <-> TF for composite models (#19732) 
* First step of PT->TF for composite models

* Update the tests

* For VisionEncoderDecoderModel

* Fix

* Fix

* Add comment

* Fix

* clean up import

* Save memory

* For (TF)EncoderDecoderModel

* For (TF)EncoderDecoderModel

Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
 2022-10-21 12:40:39 +02:00
Yih-Dar 12ce2941c7
Fix docker image build (#19759) 
* Use 2 jobs for the docker image build (latest torch + DS)

* fix

* Add comment

Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
 2022-10-20 20:36:13 +02:00
Yih-Dar 15fd39ea0e
Install tf2onnx dev version (#19755) 
* pin tf2onnx<=1.12.0

* Install tf2onnx main

* Pin to a specific commit

Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
 2022-10-20 20:24:39 +02:00
Joao Gante 5ed9bd1896
TF: sample generation compatible with XLA and dynamic batch sizes (#19773) 2022-10-20 19:01:22 +01:00
Lianmin Zheng c186e816bd
[FLAX] Add dtype to embedding for gpt2 model (#18462) 
* [FLAX] Add dtype to embedding for gpt2 model

* lint
 2022-10-20 18:15:49 +02:00
Chino baa00f65ae
Fix exception thrown using MishActivation (#19739) 
* Fix exception thrown using MishActivation

* Update activations.py
 2022-10-20 09:13:35 -04:00
Rohit Gupta 2dd1b8f0c5
adding key pair dataset (#19765) 2022-10-20 09:05:49 -04:00
IMvision12 17d7aec895
Update modeling_layoutlmv3.py (#19753) 2022-10-20 13:47:17 +01:00
Nicolas Patry a40386669f
`image-segmentation` pipeline: re-enable `small_model_pt` test. (#19716) 
* Re-enable `small_model_pt`.

Re-enable `small_model_pt`.

Enabling the current test with the current values.

Debugging the values on the CI.

More logs ? Printing doesn't work ?

Using the CI values instead. Seems to be a Pillow sensitivity.

* Update src/transformers/pipelines/image_segmentation.py

Co-authored-by: Alara Dirik <8944735+alaradirik@users.noreply.github.com>

Co-authored-by: Alara Dirik <8944735+alaradirik@users.noreply.github.com>
 2022-10-20 11:57:11 +02:00
Rohith ND eb98da9880
[Doctest] OpenAIGPTConfig and OPTConfig (#19763) 2022-10-20 10:22:00 +02:00