Li-Huai (Allan) Lin
5c702175eb
up ( #18805 )
2022-08-30 12:30:46 +02:00
Ekagra Ranjan
da02b4035c
Add docstring for BartForCausalLM ( #18795 )
...
* add docstring for BartForCausalLM
* doc-style fic
2022-08-30 12:19:03 +02:00
amyeroberts
8c4a11493f
Revert to and safely handle flag in owlvit config ( #18750 )
2022-08-29 18:48:24 +02:00
Yih-Dar
da5bb29219
send model to the correct device ( #18800 )
...
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
2022-08-29 18:46:30 +02:00
NielsRogge
f1fd460694
Add SegFormer and ViLT links ( #18808 )
...
Co-authored-by: Niels Rogge <nielsrogge@Nielss-MacBook-Pro.local>
2022-08-29 18:46:07 +02:00
Lucain
169b8cde47
Fix mock in test_cached_files_are_used_when_internet_is_down ( #18804 )
2022-08-29 15:56:08 +02:00
Yih-Dar
8b67f20935
Fix memory leak issue in torch_fx tests ( #18547 )
...
Co-authored-by: Lysandre Debut <hi@lysand.re>
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
2022-08-29 11:43:20 +02:00
fatih
b10a3b3760
fix a possible typo in auto feature extraction ( #18779 )
2022-08-29 11:24:53 +02:00
Yih-Dar
5f06a09b9f
fix missing block when there is no failure ( #18775 )
...
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
2022-08-29 09:10:13 +02:00
Philipp Schmid
f2fbe44753
Fix broken link DeepSpeed documentation link ( #18783 )
...
* Fix broken link
* Trigger CI
Co-authored-by: Stas Bekman <stas@stason.org>
2022-08-28 19:32:19 -07:00
Duong A. Nguyen
21f6f58721
Fix incomplete outputs of FlaxBert ( #18772 )
...
* Fix incomplete FlaxBert outputs
* fix big_bird electra roberta
2022-08-26 21:04:18 +02:00
Patrick von Platen
62ceb4d661
[Wav2vec2 + LM Test] Improve wav2vec2 with lm tests and make torch version dependent for now ( #18749 )
...
* add first generation tutorial
* remove generation
* make version dependent expected values
* Apply suggestions from code review
* Update tests/models/wav2vec2_with_lm/test_processor_wav2vec2_with_lm.py
* fix typo
2022-08-26 14:11:55 +02:00
Patrick von Platen
8869bf41fe
[VisionEncoderDecoder] Add gradient checkpointing ( #18697 )
...
* add first generation tutorial
* VisionEnocderDecoder gradient checkpointing
* remove generation
* add tests
2022-08-26 14:11:27 +02:00
Joao Gante
06a6a4bd51
CLI: Improved error control and updated hub requirement ( #18752 )
2022-08-25 17:08:05 +01:00
Rahul A R
e9442440fc
streamlining 'checkpointing_steps' parsing ( #18755 )
2022-08-25 11:00:38 -04:00
Craig Chan
fbf382c84d
Determine framework automatically before ONNX export ( #18615 )
...
* Automatic detection for framework to use when exporting to ONNX
* Log message change
* Incorporating PR comments, adding unit test
* Adding tf for pip install for run_tests_onnxruntime CI
* Restoring past changes to circleci yaml and test_onnx_v2.py, tests moved to tests/onnx/test_features.py
* Fixup
* Adding test to fetcher
* Updating circleci config to log more
* Changing test class name
* Comment typo fix in tests/onnx/test_features.py
Co-authored-by: lewtun <lewis.c.tunstall@gmail.com>
* Moving torch_str/tf_str to self.framework_pt/tf
* Remove -rA flag in circleci config
Co-authored-by: lewtun <lewis.c.tunstall@gmail.com>
2022-08-25 16:31:34 +02:00
Patrick Deutschmann
3223d49354
Add ONNX support for Longformer ( #17176 )
...
* Implement ONNX support for Longformer
Fix repo consistency check complaints
Fix value mismatches
Add pooler output for default model
Increase validation atol to accommodate multiple-choice error
Fix copies
Fix chunking for longer sequence lengths
Add future comment
* Fix issue in mask_invalid_locations
* Remove torch imports in configuration_longformer
* Change config access to fix LED
* Push opset version to support tril
* Work in review comments (mostly style)
* Add Longformer to ONNX tests
2022-08-25 08:34:42 +02:00
Rahul A R
c55d6e4e10
examples/run_summarization_no_trainer: fixed incorrect param to hasattr ( #18720 )
...
* fixed incorrect param to hasattr
* simplified condition checks
* code cleanup
2022-08-24 12:12:42 -04:00
SaulLu
6667b0d7bf
add warning to let the user know that the __call__ method is faster than encode + pad for a fast tokenizer ( #18693 )
...
* add warning to let the user know that the method is slower that for a fast tokenizer
* user warnings
* fix layoutlmv2
* fix layout*
* change warnings into logger.warning
2022-08-24 06:27:56 -04:00
Juyoung Kim
dcff504e18
fixed docstring typos ( #18739 )
...
* fixed docstring typos
* Added missing colon
Co-authored-by: 김주영 <juyoung@zezedu.com>
2022-08-24 06:20:27 -04:00
dependabot[bot]
e49c71fc4c
Bump nbconvert from 6.3.0 to 6.5.1 in /examples/research_projects/lxmert ( #18742 )
...
Bumps [nbconvert](https://github.com/jupyter/nbconvert ) from 6.3.0 to 6.5.1.
- [Release notes](https://github.com/jupyter/nbconvert/releases )
- [Commits](https://github.com/jupyter/nbconvert/compare/6.3.0...6.5.1 )
---
updated-dependencies:
- dependency-name: nbconvert
dependency-type: direct:production
...
Signed-off-by: dependabot[bot] <support@github.com>
Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2022-08-24 06:12:56 -04:00
dependabot[bot]
5b24949669
Bump nbconvert in /examples/research_projects/visual_bert ( #18741 )
...
Bumps [nbconvert](https://github.com/jupyter/nbconvert ) from 6.3.0 to 6.5.1.
- [Release notes](https://github.com/jupyter/nbconvert/releases )
- [Commits](https://github.com/jupyter/nbconvert/compare/6.3.0...6.5.1 )
---
updated-dependencies:
- dependency-name: nbconvert
dependency-type: direct:production
...
Signed-off-by: dependabot[bot] <support@github.com>
Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2022-08-24 06:12:48 -04:00
Daniel Stancl
c72d7d91bf
Add TF implementation of XGLMModel ( #16543 )
...
* Add TFXGLM models
* Add todo: self.supports_xla_generation = False
Co-authored-by: Daniel Stancl <stancld@Daniels-MacBook-Pro.local>
Co-authored-by: Daniel Stancl <stancld@daniels-mbp.home>
Co-authored-by: Joao Gante <joaofranciscocardosogante@gmail.com>
Co-authored-by: Daniel <daniel.stancl@rossum.ai>
Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>
2022-08-24 10:51:05 +01:00
Yih-Dar
cecf9f9b27
fix pipeline_tutorial.mdx doctest ( #18717 )
...
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
2022-08-24 05:38:03 -04:00
Constantin Hütterer
a442884b87
Add minor doc-string change to include hp_name param in hyperparameter_search ( #18700 )
...
* Add minor doc-string change to include hp_name
* fix: missing type-information for kwargs
* fix: missing white-space in hyperparameter_search doc-strings
2022-08-24 05:07:17 -04:00
Mishig Davaadorj
c12dbdc246
Update perf_infer_gpu_many.mdx ( #18744 )
2022-08-24 10:37:52 +02:00
Joao Gante
6faf283288
CLI: Don't check the model head when there is no model head ( #18733 )
2022-08-23 15:38:59 +01:00
SaulLu
438698085c
improve add_tokens docstring ( #18687 )
...
* improve add_tokens documentation
* format
2022-08-23 07:23:51 -04:00
Nicolas Patry
891704b3c2
Removing warning of model type for microsoft/tapex-base-finetuned-wtq ( #18711 )
...
and friends.
2022-08-23 13:17:06 +02:00
Yih-Dar
84beb8a49b
Unpin detectron2 ( #18727 )
...
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
2022-08-23 11:10:07 +02:00
Atharva Ingle
d90a36d192
remove check for main process for trackers initialization ( #18706 )
2022-08-22 11:16:27 -04:00
tgadeliya
0f257a8774
Add missing tokenizer tests - Longformer ( #17677 )
2022-08-22 12:13:20 +02:00
Yih-Dar
3fa45dbd91
Fix Data2VecVision ONNX test ( #18587 )
...
Co-authored-by: lewtun <lewis.c.tunstall@gmail.com>
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
2022-08-22 11:28:23 +02:00
Yih-Dar
30992ef0d9
[Hotfix] pin detectron2 5aeb252 to avoid test fix ( #18701 )
...
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
2022-08-20 00:37:38 +02:00
Patrick von Platen
1f3c2282b5
Temp fix for broken detectron2 import ( #18699 )
...
* add first generation tutorial
* [Circle CI] Temporary fix for broken detectron2 import
* remove generation
2022-08-19 22:55:33 +02:00
Joao Gante
e95d433d77
Generate: add missing **model_kwargs in sample tests ( #18696 )
2022-08-19 16:14:27 +01:00
Atharva Ingle
e54a1b49aa
model.tie_weights() should be applied after accelerator.prepare() (#18676 )
...
* `model.tie_weights()` should be applied after `accelerator.prepare`
Weight tying should be done after the model has been moved to XLA device as mentioned on PyTorch/XLA Troubleshooting guide [here](https://github.com/pytorch/xla/blob/master/TROUBLESHOOTING.md#xla-tensor-quirks )
* format code
2022-08-18 13:46:57 -04:00
Loubna Ben Allal
bbbb453e58
Add an examples folder for code downstream tasks ( #18679 )
...
* add examples subfolder
* mention examples in codeparrot readme
* use Trainer optimizer and scheduler type and add output_dir as argument
* add example of text-to-python and python-to-text models
* mention the downstream examples in the readme
* fix typo
2022-08-18 18:24:24 +02:00
Younes Belkada
a123eee9df
[bnb] Move documentation ( #18671 )
...
* fix bnb documentation
- move bnb documentation to `infer_gpu_many`
* small refactoring
- added text on infer_gpu_one
- added a small note on infer_gpu_many
- added customized multi gpu example on infer_gpu_many
* Update docs/source/en/perf_infer_gpu_many.mdx
Co-authored-by: Stas Bekman <stas00@users.noreply.github.com>
* apply suggestions
Co-authored-by: Stas Bekman <stas00@users.noreply.github.com>
* Apply suggestions from code review
Co-authored-by: Stas Bekman <stas00@users.noreply.github.com>
Co-authored-by: Stas Bekman <stas00@users.noreply.github.com>
2022-08-18 17:34:48 +02:00
Zachary Mueller
358fc18613
Add evaluate to examples requirements ( #18666 )
2022-08-18 10:57:39 -04:00
Severin Simmler
d243112b65
Fix breaking change in onnxruntime for ONNX quantization ( #18336 )
...
* Fix quantization
* Save model
* Remove unused comments
* Fix formatting
2022-08-18 10:06:16 -04:00
lewtun
5987c637ee
Fix repo consistency ( #18682 )
2022-08-18 09:47:50 -04:00
regisss
76454b08c8
Rename second input dimension from "sequence" to "num_channels" for CV models ( #17976 )
2022-08-18 15:13:54 +02:00
amyeroberts
780253ce3d
Rename method to avoid clash with property ( #18677 )
2022-08-18 12:56:27 +01:00
Yih-Dar
2c947d2939
Ping detectron2 for CircleCI tests ( #18680 )
...
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
2022-08-18 12:57:18 +02:00
Joao Gante
a541d97477
Generate: validate model_kwargs on FLAX (and catch typos in generate arguments) ( #18653 )
2022-08-18 10:56:21 +01:00
Patrick von Platen
0ea53822f8
[LongT5] Correct docs long t5 ( #18669 )
...
* add first generation tutorial
* [LongT5 Docs] Correct docs
* correct expected string
* remove incorrect file
2022-08-18 10:03:50 +02:00
Matt
582c537175
Allow users to force TF availability ( #18650 )
...
* Allow users to force TF availability
* Correctly name the envvar!
2022-08-18 03:09:09 -04:00
amyeroberts
49e44b216b
Update feature extractor methods to enable type cast before normalize ( #18499 )
...
* Update methods to optionally rescale
This is necessary to allow for casting our images / videos to numpy arrays within the feature extractors' call. We want to do this to make sure the behaviour is as expected when flags like are False. If some transformations aren't applied, then the output type can't be unexpected e.g. a list of PIL images instead of numpy arrays.
* Cast images to numpy arrays in call to enable consistent behaviour with different configs
* Remove accidental clip changes
* Update tests to reflect the scaling logic
We write a generic function to handle rescaling of our arrays. In order for the API to be intuitive, we take some factor c and rescale the image values by that. This means, the rescaling done in normalize and to_numpy_array are now done with array * (1/255) instead of array / 255. This leads to small differences in the resulting image. When testing, this was in the order of 1e-8, and so deemed OK
2022-08-17 19:57:07 +01:00
Jingya HUANG
86d0b26d6c
Fix matmul inputs dtype ( #18585 )
2022-08-17 15:59:43 +02:00
