Commits · revert-17547-update-support-image · 某某某 / transformers-new

03 Jun, 2022 9 commits

Revert "Update index.mdx (#17547)" · 47727e36
Britney Muller authored 3 years ago
```
This reverts commit 72f5b949.
```
47727e36

Britney Muller authored 3 years ago

This PR updates our Expert Acceleration Program image with a new image featuring our experts.

This is similar to our Transformers/README.md image update that has proven to be successful.

72f5b949

Clean imports to fix test_fetcher (#17531) · c4e58cd8

Sylvain Gugger authored 3 years ago


* Clean imports to fix test_fetcher

* Add dependencies printer

* Update utils/tests_fetcher.py

Co-authored-by: lewtun <lewis.c.tunstall@gmail.com>

* Fix Perceiver import

Co-authored-by: lewtun <lewis.c.tunstall@gmail.com>

c4e58cd8

Update run_glue_no_trainer.py (#17546) · 254d9c06
bhuang authored 3 years ago

254d9c06
Fix all offload and MP tests (#17533) · 83439012
Sylvain Gugger authored 3 years ago

83439012
Fix bug - layer names and activation from previous refactor (#17524) · 1c57242d
amyeroberts authored 3 years ago
```
* Fix activation and layers in MLP head

* Remove unused import
```
1c57242d

Add support for Perceiver ONNX export (#17213) · babeff55

Patrick Deutschmann authored 3 years ago


* Start adding perceiver support for ONNX

* Fix pad token bug for fast tokenizers

* Fix formatting

* Make get_preprocesor more opinionated (processor priority, otherwise tokenizer/feature extractor)

* Clean docs format

* Minor cleanup following @sgugger's comments

* Fix typo in docs

* Fix another docs typo

* Fix one more typo in docs

* Update src/transformers/onnx/utils.py

Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* Update src/transformers/onnx/utils.py

Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* Update src/transformers/onnx/utils.py

Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

babeff55

Allow from transformers import TypicalLogitsWarper (#17477) · 5c17918f

Robert Dargavel Smith authored 3 years ago

* Allow from transformers import TypicalLogitsWarper

* Added TypicalLogitsWarper

* Allow from transformers import TypicalLogitsWarper

* Allow from transformers import TypicalLogitsWarper

* Allow from transformers import TypicalLogitsWarper

* Allow from transformers import TypicalLogitsWarper

Added TypicalLogitsWarper

Allow from transformers import TypicalLogitsWarper

Allow from transformers import TypicalLogitsWarper

Allow from transformers import TypicalLogitsWarper

5c17918f

Add Gated-SiLU to T5 (#17420) · 607acd4f

DanielHesslow authored 3 years ago


* Add gated-silu to t5 architecture to support UL2

* Fix error message

* formatting

* formatting again

* refactor

* fix classnames in _init_weights

* remove is_gated

* add test

* fix test

* Try without the test?

* Add back the test.

* Improve error message.

Co-authored-by: Daniel Hesslow <daniel@lighton.ai>

607acd4f

02 Jun, 2022 11 commits
- Update URL for Hub PR docs (#17532) · 1c220ced
  lewtun authored 3 years ago
  
  1c220ced
- fix OPT-Flax CI tests (#17512) · 013462c5
  Arthur authored 3 years ago
  
  013462c5
- [trainer/deepspeed] load_best_model (reimplement re-init) (#17151) · 2f59ad16
  Stas Bekman authored 3 years ago
```
* [trainer/deepspeed] load_best_model

* to sync with DS PR #1947

* simplify

* rework load_best_model test

* cleanup

* bump deepspeed>=0.6.5

Co-authored-by: Olatunji Ruwase <olruwase@microsoft.com>
```
  2f59ad16
- Implemented loss for training AudioFrameClassification (#17513) · 046c5ea9
  Moreno La Quatra authored 3 years ago
```
* Implemented loss for training AudioFrameClassification

* reported changes in wav2vec2 main class and used make copies to propagate

* running black for code formatting
```
  046c5ea9
- Update configuration_auto.py (#17527) · 085321c9
  Kamal Raj authored 3 years ago
  
  085321c9
- Check list of models in the main README and sort it (#17517) · 048dd73b
  Sylvain Gugger authored 3 years ago
```
* Script for README

* Fix copies

* Complete error message
```
  048dd73b
- Fix when Accelerate is not installed (#17518) · 588d8f1f
  Sylvain Gugger authored 3 years ago
  
  588d8f1f
- Clean README in post release job as well. (#17519) · f128ccb9
  Sylvain Gugger authored 3 years ago
  
  f128ccb9
- Fix CI tests hang forever (#17471) · 216499bf
  Yih-Dar authored 3 years ago
```
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
```
  216499bf
- Print more library versions in CI (#17384) · 659b27fd
  Yih-Dar authored 3 years ago
```
* print more lib. versions and just befor test runs

* update print_env_pt.py

* rename to print_env

* Disable warning + better job name

* print python version

Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
```
  659b27fd
- Split push CI into 2 workflows (#17369) · 0932adb3
  Yih-Dar authored 3 years ago
```
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
```
  0932adb3
01 Jun, 2022 17 commits

Fix Tapas tests (#17510) · 58fb3c9f
Yih-Dar authored 3 years ago
```
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
```
58fb3c9f
CLI: tool to convert PT into TF weights and open hub PR (#17497) · ca1f1c86
Joao Gante authored 3 years ago

ca1f1c86
Fix flakey no-trainer test (#17515) · 3766df4f
Zachary Mueller authored 3 years ago

3766df4f
Deal with the error when task is regression (#16330) · 028d4b7c
fireindark707 authored 3 years ago

028d4b7c

Adding LeViT Model by Facebook (#17466) · 84aaadd8

Anugunj Naman authored 3 years ago


* levit files

* levit tests

* weights script

* weights script

* update

* style fixes

* few minor corrections

* Added teacher model

* edit docs

* fix-copies

* style fixes

* pr error resolved

* Update README.md

Co-authored-by: NielsRogge <48327001+NielsRogge@users.noreply.github.com>

* Update docs/source/en/index.mdx

Co-authored-by: NielsRogge <48327001+NielsRogge@users.noreply.github.com>

* Update docs/source/en/model_doc/levit.mdx

Co-authored-by: NielsRogge <48327001+NielsRogge@users.noreply.github.com>

* Update docs/source/en/model_doc/levit.mdx

Co-authored-by: NielsRogge <48327001+NielsRogge@users.noreply.github.com>

* Update docs/source/en/model_doc/levit.mdx

Co-authored-by: NielsRogge <48327001+NielsRogge@users.noreply.github.com>

* Update docs/source/en/model_doc/levit.mdx

Co-authored-by: NielsRogge <48327001+NielsRogge@users.noreply.github.com>

* Update src/transformers/__init__.py

Co-authored-by: NielsRogge <48327001+NielsRogge@users.noreply.github.com>

* Update src/transformers/models/levit/__init__.py

Co-authored-by: NielsRogge <48327001+NielsRogge@users.noreply.github.com>

* Update src/transformers/models/levit/configuration_levit.py

Co-authored-by: NielsRogge <48327001+NielsRogge@users.noreply.github.com>

* Update src/transformers/models/levit/configuration_levit.py

Co-authored-by: NielsRogge <48327001+NielsRogge@users.noreply.github.com>

* Update src/transformers/models/levit/feature_extraction_levit.py

Co-authored-by: NielsRogge <48327001+NielsRogge@users.noreply.github.com>

* suggested pr changes

* style fixes

* minor bug

* update

* minor doc edit

* style

* Update src/transformers/models/levit/feature_extraction_levit.py

Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* Update src/transformers/models/levit/feature_extraction_levit.py

Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* Update tests/models/levit/test_modeling_levit.py

Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* Update src/transformers/models/levit/modeling_levit.py

Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* Update src/transformers/models/levit/feature_extraction_levit.py

Co-authored-by: NielsRogge <48327001+NielsRogge@users.noreply.github.com>

* residual layer readable

* style

* Update docs/source/en/model_doc/levit.mdx

Co-authored-by: NielsRogge <48327001+NielsRogge@users.noreply.github.com>

* Update src/transformers/models/levit/feature_extraction_levit.py

Co-authored-by: NielsRogge <48327001+NielsRogge@users.noreply.github.com>

* Update src/transformers/models/levit/feature_extraction_levit.py

Co-authored-by: NielsRogge <48327001+NielsRogge@users.noreply.github.com>

* Update src/transformers/models/levit/feature_extraction_levit.py

Co-authored-by: NielsRogge <48327001+NielsRogge@users.noreply.github.com>

* Update src/transformers/models/levit/feature_extraction_levit.py

Co-authored-by: NielsRogge <48327001+NielsRogge@users.noreply.github.com>

* Update src/transformers/models/levit/modeling_levit.py

Co-authored-by: NielsRogge <48327001+NielsRogge@users.noreply.github.com>

* Update src/transformers/models/levit/modeling_levit.py

Co-authored-by: NielsRogge <48327001+NielsRogge@users.noreply.github.com>

* Update src/transformers/models/levit/modeling_levit.py

Co-authored-by: NielsRogge <48327001+NielsRogge@users.noreply.github.com>

* Update tests/models/levit/test_feature_extraction_levit.py

Co-authored-by: NielsRogge <48327001+NielsRogge@users.noreply.github.com>

* change checkpoints and style

* update

* minor changes

* Update src/transformers/models/levit/modeling_levit.py

Co-authored-by: NielsRogge <48327001+NielsRogge@users.noreply.github.com>

* Update src/transformers/models/levit/modeling_levit.py

Co-authored-by: NielsRogge <48327001+NielsRogge@users.noreply.github.com>

Co-authored-by: NielsRogge <48327001+NielsRogge@users.noreply.github.com>
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

84aaadd8

Fix CTRL tests (#17508) · 1d2b57b8

Yih-Dar authored 3 years ago


* fix

Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>

1d2b57b8

Fix LayoutXLMProcessorTest (#17506) · 693720e5
Yih-Dar authored 3 years ago
```
* fix

Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
```
693720e5

Debug LukeForMaskedLM (#17499) · 4d1ce396

Ryokan RI authored 3 years ago

* add a test for a word only input

* make LukeForMaskedLM work without entity inputs

* update test

* add LukeForMaskedLM to MODEL_FOR_MASKED_LM_MAPPING_NAMES

* restore pyproject.toml

* empty line at the end of pyproject.toml

4d1ce396

Fix MP and CPU offload tests for Funnel and GPT-Neo (#17503) · 4390151b
Sylvain Gugger authored 3 years ago

4390151b
Exclude Databricks from notebook env (#17496) · 6813439f
Sylvain Gugger authored 3 years ago

6813439f

Fix `tokenizer` type annotation in `pipeline(...)` (#17500) · 3042ea4f

Will Frey authored 3 years ago

I think you mean to accept either an instance of `PreTrainedTokenizer` or `PreTrainedTokenizerFast` inside of the `pipeline(...)` factory function, if the `tokenizer` argument isn't a `str`.

3042ea4f

Refactor classes to inherit from nn.Module instead of nn.Sequential (#17493) · bdc01711
amyeroberts authored 3 years ago
```
* Adapt Maskformer, VAN, ResNet and RegNet modules to inherit from nn.Module
```
bdc01711
Fix wav2vec2 export onnx model with attention_mask error (#16004) · b1160c0b
nilboy authored 3 years ago
```
* Fix wav2vec2 export onnx model with attention_mask error

* fix repository_consistency
```
b1160c0b

Add warning when using older version of torch for ViltFeatureExtractor (#16756) · d91da4c6

Xing Han Lu authored 3 years ago


* Update feature_extraction_vilt.py

* apply black

* Update imports

* Change warning to logging

* Use logger instead of logging.logging

* make fixup

* Move error message

* Update src/transformers/models/vilt/feature_extraction_vilt.py

Co-authored-by: Xing Han Lu <xhlperso@gmail.com>
Co-authored-by: Lysandre Debut <lysandre@huggingface.co>

d91da4c6

Fix typo of variable names for key and query projection layer (#17155) · 24092b14

Kyeongpil Kang authored 3 years ago

self.pos_proj and self.pos_q_proj should be changed to self.pos_key_proj and self.pos_query_proj as same as PyTorch implements.

24092b14

Fixed wrong error message for missing weight file (#17216) · 811da2b8
Jimin Park authored 3 years ago

811da2b8

Add OnnxConfig for SqueezeBert iss17314 (#17315) · 4f38808e

Ruihua Fang authored 3 years ago


* add onnx config for SqueezeBert

* add test for onnx config for SqueezeBert

* add automatically updated doc for onnx config for SqueezeBert

* Update src/transformers/onnx/features.py

Co-authored-by: lewtun <lewis.c.tunstall@gmail.com>

* Update src/transformers/models/squeezebert/configuration_squeezebert.py

Co-authored-by: lewtun <lewis.c.tunstall@gmail.com>

Co-authored-by: lewtun <lewis.c.tunstall@gmail.com>

4f38808e

31 May, 2022 3 commits

[GPT2Tokenizer] Fix GPT2 with bos token (#17498) · ba286fe7
Patrick von Platen authored 3 years ago

ba286fe7

Opt in flax and tf (#17388) · 7822a9b7

Arthur authored 3 years ago


* initial commit

* add init file

* update globakl init

* update index and dummy objects

* style

* update modelling auto

* fix initi typo in src/transformers

* fix typo in modeling tf auto, opt was in wrong mapping name

* fixed a slow test : saved_model

* style

* fix positionnal embedding if no position id is provided

* update tf test

* update test flax requirements

* fixed serialization

* update

* update tf name to allow smooth convertion

* update flax tests

* style

* fix test typo

* fix tf typo test

* add xla for generate support in causal LM

* fixed bug

* cleaned tf tests

* style

* removed from PT for slow tests

* fix typp

* opt test as slow

* trying to fix GPT2 undefined

* correct documentation and add to test doc

* update tf doc

* fix doc

* fake commit

* Apply suggestions from code review

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

* update test based on review

* merged main layer for functionning test

* fixup + quality

* Apply suggestions from code review

Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* update long comment

* make fix copies

Co-authored-by: Arthur <arthur@huggingface.co>
Co-authored-by: Joao Gante <joaofranciscocardosogante@gmail.com>
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

7822a9b7

[Json configs] Make json prettier for all saved tokenizer files & ensure same... · f394a2a5

Patrick von Platen authored 3 years ago

[Json configs] Make json prettier for all saved tokenizer files & ensure same json format for all processors (tok + feat_extract) (#17457)

* [Json dump] Make json prettier

* correct more tokenizeirs

* more patterns

* add aggressive test

* the aggressive test was actually useful :-)

* more tests

* Apply suggestions from code review

f394a2a5