Commit Graph

12327 Commits

Author SHA1 Message Date
Sylvain Gugger
b45192ec47
Fix big model inference for T5 models in float16 (#22095)
* Fix big model inference for T5 models in float16

* Apply suggestions from code review

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

* Style

* Trigger CI with latest release

---------

Co-authored-by: Younes Belkada <49240599+younesbelkada@users.noreply.github.com>
2023-03-14 09:20:16 -04:00
Nicola Procopio
7f5ad6c35b
Translation Italian: perf_train_cpu and perf_train_cpu_many (#22151)
* added translated files

added perf_train_cpu and perf_train_cpu_many

* updated toctree
2023-03-14 11:09:36 +00:00
Yih-Dar
ff88703501
Update 2 doctest expected values for torch 2.0.0 (#22148)
update values

Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
2023-03-14 09:13:16 +00:00
Alara Dirik
cdddfbffa1
Add ConvNeXT V2 (#21679)
* Add ConvNeXt V2 to transformers
* TF model is separated from the PR to fix issues
2023-03-14 12:08:14 +03:00
Yih-Dar
6c2ad00c46
Move is_pipeline_test_to_skip to specific model test classes (#21999)
* Move `is_pipeline_test_to_skip` to specific model test classes

---------

Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
2023-03-14 10:03:02 +01:00
Arthur
2beabd24f0
[🛠️] Fix-whisper-breaking-changes (#21965)
* temp fix

* temporary fix

* update

* fix tests

* fixup

* update based on reveiew

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

* update to fix tests

* update docstring

---------

Co-authored-by: Sanchit Gandhi <93869735+sanchit-gandhi@users.noreply.github.com>
2023-03-14 09:23:48 +01:00
MichaelRipa
101a6cd276
docs: New terms and updates to glossary (#21982)
* Updated glossary with new terms, added abbreviations for certain terms and merged autoencoding models, autoregressive models and causal language modeling into encoder and decoder models

* Update docs/source/en/glossary.mdx

Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>

* Update docs/source/en/glossary.mdx

Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>

* Update docs/source/en/glossary.mdx

Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>

* Update docs/source/en/glossary.mdx

Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>

* Update docs/source/en/glossary.mdx

Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>

* Update docs/source/en/glossary.mdx

Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>

* Update docs/source/en/glossary.mdx

Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>

* Update docs/source/en/glossary.mdx

Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>

* Update docs/source/en/glossary.mdx

Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>

* Update docs/source/en/glossary.mdx

Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>

* Update docs/source/en/glossary.mdx

Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>

* Update docs/source/en/glossary.mdx

Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>

* Added link to 'Pipeline for inference' tutorial

* Trigger CI

* Update docs/source/en/glossary.mdx

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

* Update docs/source/en/glossary.mdx

Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>

* Added entry for self supervised learning, added deleted entries + fixed broken links

* Update docs/source/en/glossary.mdx

Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>

---------

Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>
Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>
2023-03-13 19:09:37 -04:00
Yih-Dar
ba9e0191de
Prepare daily CI for torch 2.0.0 (#22135)
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
2023-03-13 22:21:15 +01:00
Patrick von Platen
f780557a34
[Safetensors] Add explicit flag to from pretrained (#22083)
* [Safetensors] Add explicit  flag to from pretrained

* add test

* remove @

* Apply suggestions from code review

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

---------

Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>
2023-03-13 21:39:06 +01:00
Sylvain Gugger
3a35937ede
Remove backend check for torch.compile (#22140)
* Remove backend enforcment for torch.compile

* Update error

* Update src/transformers/training_args.py

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>

* Style

---------

Co-authored-by: Stas Bekman <stas00@users.noreply.github.com>
2023-03-13 16:34:00 -04:00
Stas Bekman
618697ef53
[deepspeed docs] Activation Checkpointing (#22099)
* [deepspeed docs] Activation Checkpointing

* Apply suggestions from code review

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

* Update deepspeed.mdx

---------

Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>
2023-03-13 12:52:42 -07:00
Stas Bekman
5b85add7d5
[trainer] fix bug in grad accum with multiple epochs (#22098)
* [trainer] fix bug in grad accum

* comment out debug

* fix one-off

* rename counter
2023-03-13 12:51:40 -07:00
Sylvain Gugger
1c801d65eb
Enforce same behavior as PyTorch 2.0 for older versions (#22136) 2023-03-13 15:50:50 -04:00
Joao Gante
e16cbe88ae
Trainer: let generate pick its inputs (#22108)
* Let generate pick its inputs

* fix squad seq2seq example
2023-03-13 19:00:25 +00:00
Younes Belkada
d979cf6efd
[Whiper] add get_input_embeddings to WhisperForAudioClassification (#22133)
* add `get_input_embeddings` to `WhisperForAudioClassification`

* add common tests

* fix another common test

* Update tests/models/whisper/test_modeling_whisper.py

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

* fix style

---------

Co-authored-by: Arthur <48595927+ArthurZucker@users.noreply.github.com>
2023-03-13 19:46:01 +01:00
bishmdl76
987972377d
Update configuration_align.py (projected_dim=640) (#22139)
Update configuration_align.py

updated projected_dim=640 from 512 in arguments of AlignConfig
2023-03-13 14:12:12 -04:00
Yih-Dar
54ee56b15b
Add a new script to check model testers' config (#22063)
* Add script

---------

Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
2023-03-13 19:11:19 +01:00
mollerup23
a096eaca65
Adding Type Hints to TF_Pegasus model (#21941)
* Adding Type Hints to TF_Pegasus model

* Updated some parameters per maintainer comments
2023-03-13 15:58:29 +00:00
Sylvain Gugger
6cb5132a7f
Fix doc link for MGP-STR (#22138) 2023-03-13 15:26:50 +00:00
Maria Khalusova
8def252de2
Zero-shot image classification task guide (#22132)
* WIP

* WIP

* manual inference example

* make style

* Apply suggestions from code review

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

---------

Co-authored-by: Alara Dirik <8944735+alaradirik@users.noreply.github.com>
2023-03-13 10:57:17 -04:00
Karim Foda
e61081e725
Fix gradient checkpointing bug in trocr (#22126)
* Fix gradient checkpointing bug in trocr

* Fix format

* Update src/transformers/models/trocr/modeling_trocr.py

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

---------

Co-authored-by: Younes Belkada <49240599+younesbelkada@users.noreply.github.com>
2023-03-13 15:45:47 +01:00
Karim Foda
ef74e7e783
Fix gradient checkpointing bug in LongT5 (#22130) 2023-03-13 14:06:17 +00:00
Karim Foda
c1db6a3bab
Fix gradient checkpointing bug in xmod (#22129) 2023-03-13 15:05:11 +01:00
Younes Belkada
6652e7da0d
[Blip2] skip accelerate test (#22124)
skip accelerate test
2023-03-13 15:03:21 +01:00
Nicola Procopio
dd3a0580a6
Added big_models.mdx italian translation #17600 (#22115)
* updated toctree

* italian translation big_model.mdx

* italian translation big_models
2023-03-13 10:02:03 -04:00
Karim Foda
0768c5e274
Fix gradient checkpointing bug in xlm_roberta_xl (#22128) 2023-03-13 13:52:34 +00:00
Karim Foda
4c14c1f47b
Fix gradient checkpointing bug in Trajectory Transformer (#22125) 2023-03-13 13:50:02 +00:00
Karim Foda
d0876a095f
Fix gradient checkpointing bug in xglm (#22127) 2023-03-13 13:49:23 +00:00
Alex Calabrese
0c883766bd
Add pr_checks.mdx Italian translation (#17459) (#22116)
* Add pr_checks.mdx Italian translation (#17459)

* Updated pr_checks.mdx Italian translation (#17459)
2023-03-13 09:24:34 -04:00
wangpeng
102b5ff4a8
add new model of MGP-STR (#21418)
* add new model of MGP-STR

* fix the check failings

* remove torch and numpy from mgp_tokenization

* remove unused import from modeling_mgp_str

* add test_processing_mgp_str

* rm test_processing_mgp_str.py

* add test_processing_mgp_str

* add test_processing_mgp_str

* add test_processing_mgp_str

* rm test_processing_mgp_str and add softmax outs to model

* rm test_processing_mgp_str and add softmax outs to model

* rewrite the code of mgp-str according to PR suggestions

* rewrite the code of mgp-str according to PR suggestions

* add new model of MGP-STR

* fix the check failings

* remove torch and numpy from mgp_tokenization

* remove unused import from modeling_mgp_str

* add test_processing_mgp_str

* rm test_processing_mgp_str.py

* add test_processing_mgp_str

* add test_processing_mgp_str

* add test_processing_mgp_str

* rm test_processing_mgp_str and add softmax outs to model

* rewrite the code of mgp-str according to PR suggestions

* rewrite the code of mgp-str according to PR suggestions

* remove representation_size from MGPSTRConfig

* reformat configuration_mgp_str.py

* format test_processor_mgp_str.py

* add test for tokenizer and complete model/processer test and model file

* rm Unnecessary tupple in modeling_mgp_str

* reduce hidden_size/layers/label_size in test_model

* add integration tests and change MGPSTR to Mgpstr

* add test for logit values

* reformat test model file

---------

Co-authored-by: yue kun <yuekun.wp@alibaba-inc.com>
2023-03-13 10:11:31 +00:00
Alara Dirik
32e3466d38
Add AutoModelForZeroShotImageClassification (#22087)
Adds AutoModelForZeroShotImageClassification to transformers
2023-03-13 12:46:14 +03:00
Sanchit Gandhi
b90fbc7e0b
[Whisper] Remove embed_tokens from encoder docstring (#21996)
* [Whisper] Remove embed_tokens from encoder docstring

* new line to retrigger CI

* remove new line
2023-03-11 14:03:36 +01:00
Yih-Dar
2f320661f3
Revert "[GPT2] Propose fix for #21080" (#22093)
Revert "[GPT2] Propose fix for #21080 (#21853)" to avoid CI failure

This reverts commit a3fef89b26.
2023-03-10 22:08:21 +01:00
Sylvain Gugger
499770c088
Fix imports of TF MobileViT (#22065)
* Fix imports of TF MobileViT

* Fix copies
2023-03-10 14:46:34 -05:00
Maria Khalusova
bdec2768bd
GPT-J specific half precision on CPU note (#22086)
* re: #21989

* update re: #21989

* removed cpu option

* make style
2023-03-10 14:03:43 -05:00
Dean Wyatte
2f4cdd97f5
handle numpy inputs in whole word mask data collator (#22032) 2023-03-10 10:50:29 -05:00
J-shang
a70da86b84
Fix hint in src/transformers/modeling_utils.py (#22074)
fix hint
2023-03-10 08:56:42 -05:00
Karim Foda
419d979f7f
Fix gradient checkpointing bug in Speecht5 (#22080)
* Fix gradient checkpointing bug in Speecht5

* Update modeling_speech_to_text.py

* Update src/transformers/models/speech_to_text/modeling_speech_to_text.py

* Fix change errors

---------

Co-authored-by: Joao Gante <joaofranciscocardosogante@gmail.com>
2023-03-10 13:36:09 +00:00
Joao Gante
7014fc360d
Generate - Fix broken documentation links (#22078)
fix broken links
2023-03-10 13:28:30 +00:00
Kevin Jiang
ade26bf991
Fix small typo in flan-ul2.mdx (#22068)
* Update flan-ul2.mdx

* Update flan-ul2.mdx
2023-03-10 07:44:45 -05:00
Arthur
a3fef89b26
[GPT2] Propose fix for #21080 (#21853)
* Make sure position ids are masked

* test that padded input produce the same results

* fix failing tests

* fixup

* fix batch test
2023-03-10 07:15:25 -05:00
Karim Foda
eee195b3aa
Fix gradient checkpointing bug in switch transformer (#22081) 2023-03-10 11:31:08 +00:00
Karim Foda
b9273353dc
Fix gradient checkpointing bug in Speech2Text (#22079)
* Fix gradient checkpointing bug in Speech2Text

* Update modeling_speech_to_text.py

* Update modeling_speech_to_text_2.py
2023-03-10 11:30:42 +00:00
Sylvain Gugger
a9bd5df16a
Add a progress bar for the total download of shards (#22062)
* Add a progress bar for the total download of shards

* Check for no cache at all

* Fix check
2023-03-09 16:58:03 -05:00
aws-sangeetha
1a5fc300f4
Fix case when using --gradient_accumulation_steps with DDP disabled. (#22007)
Co-authored-by: EC2 Default User <ec2-user@ip-172-31-42-72.us-west-2.compute.internal>
2023-03-09 14:31:58 -05:00
Yih-Dar
6d9031f285
Update tiny model creation script (#22058)
Update the script

Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
2023-03-09 19:53:54 +01:00
Sylvain Gugger
7a2b915e92
Add setters by type of args to TrainingArguments (#21570)
* Add setters by type of args to TrainingArguments

* Define more setters
2023-03-09 13:13:23 -05:00
Yih-Dar
ab81d31d20
Skip 3 tests for WhisperEncoderModelTest (#22060)
* skip 3 tests

---------

Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
2023-03-09 19:09:23 +01:00
Jiali Mei
8434cb878e
Edit the docstring of image_processing_donut to match code (#22033)
* Edit the docstring of `image_processing_donut` to match code

* improve style

* more style improvement after installing quality
2023-03-09 17:35:43 +00:00
Stas Bekman
ec24132b6c
[deepspeed] offload + non-cpuadam optimizer exception (#22043)
* [deepspeed] offload + non-cpuadam optimizer exception

* flip

* revert min version
2023-03-09 08:12:57 -08:00