Tanay Mehta
b8def68934
Fix Mega chunking error when using decoder-only model ( #25765 )
...
* add: potential fix to mega chunking in decoder only model bug
* add: decoder with chunking test
* add: input_mask passed with input_ids
2023-09-05 21:50:14 +02:00
Arthur
4fa0aff21e
[VITS] tokenizer integration test: fix revision did not exist ( #25996 )
...
* revision did not exist
* correct revision
2023-09-05 21:21:33 +02:00
Arthur
d0354e5e86
[CI] Fix red CI and ERROR failed should show ( #25995 )
...
* start with error too
* fix ?
* start with nit
* one more path
* use `job_name`
* mark pipeline test as slow
2023-09-05 20:16:00 +02:00
Injin Paek
6206f599e1
Add LLaMA resources ( #25859 )
...
* docs: feat: model resources for llama
* fix: resolve suggestion
Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>
Co-authored-by: Jungnerd <46880056+jungnerd@users.noreply.github.com>
Co-authored-by: Wonhyeong Seo <wonhseo@kakao.com>
---------
Co-authored-by: Steven Liu <59462357+stevhliu@users.noreply.github.com>
Co-authored-by: Jungnerd <46880056+jungnerd@users.noreply.github.com>
Co-authored-by: Wonhyeong Seo <wonhseo@kakao.com>
2023-09-05 10:50:08 -07:00
Sanchit Gandhi
8d518013ef
[Wav2Vec2 Conformer] Fix inference float16 ( #25985 )
...
* [Wav2Vec2 Conformer] Fix inference float16
* fix test
* fix test more
* clean pipe test
2023-09-05 18:26:06 +01:00
Sourab Mangrulkar
6bc517ccd4
deepspeed resume from ckpt fixes and adding support for deepspeed optimizer and HF scheduler ( #25863 )
...
* Add support for deepspeed optimizer and HF scheduler
* fix bug
* fix the import
* fix issue with deepspeed scheduler saving for hf optim + hf scheduler scenario
* fix loading of hf scheduler when loading deepspeed checkpoint
* fix import of `DeepSpeedSchedulerWrapper`
* add tests
* add the comment and skip the failing tests
* address comment
2023-09-05 22:31:20 +05:30
raghavanone
1110b565d6
Add TFDebertaV2ForMultipleChoice ( #25932 )
...
* Add TFDebertaV2ForMultipleChoice
* Import newer model in main init
* Fix import issues
* Fix copies
* Add doc
* Fix tests
* Fix copies
* Fix docstring
2023-09-05 17:13:06 +01:00
andreeahedes
da1af21dbb
PegasusX add _no_split_modules ( #25933 )
...
* no_split_modules
* no_split_modules
* inputs_embeds+pos same device
* update _no_split_modules
* update _no_split_modules
2023-09-05 16:34:34 +01:00
Abhilash Majumder
70a98024b1
Patch with accelerate xpu ( #25714 )
...
* patch with accelerate xpu
* patch with accelerate xpu
* formatting
* fix tests
* revert ruff unrelated fixes
* revert ruff unrelated fixes
* revert ruff unrelated fixes
* fix test
* review fixes
* review fixes
* black fixed
* review commits
* review commits
* style fix
* use pytorch_utils
* revert markuplm test
2023-09-05 15:41:42 +01:00
Yih-Dar
aa5c94d38d
Show failed tests on CircleCI layout in a better way ( #25895 )
...
* update
* update
* fix
---------
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
2023-09-05 15:49:33 +02:00
Joao Gante
9a70d6e56f
Trainer: delegate default generation values to generation_config ( #25987 )
2023-09-05 14:47:00 +01:00
Sahel Sharify
aea761499f
Update training_args.py to remove the runtime error ( #25920 )
...
This cl iterates through a list of keys rather than dict items while updating the dict elements. Fixes the following error:
File "..../transformers/training_args.py", line 1544, in post_init
for k, v in self.fsdp_config.items():
RuntimeError: dictionary keys changed during iteration
2023-09-05 12:43:51 +01:00
Traun Leyden
7011cd8667
Update RAG README.md with correct path to examples/seq2seq ( #25953 )
...
Update README.md with correct path to examples/seq2seq
2023-09-05 12:31:59 +01:00
Julien Chaumond
6316ce8d27
[doc] Always call it Agents for consistency ( #25958 )
2023-09-05 12:27:20 +01:00
Yih-Dar
391f26459a
Use main in conversion script ( #25973 )
...
* fix
* fix
---------
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
2023-09-05 13:04:49 +02:00
Kai
6f125aaa48
fix typo ( #25981 )
...
rename doanloading to downloading
2023-09-05 11:13:06 +01:00
Susnato Dhar
52a46dc57b
Add Pop2Piano space demo. ( #25975 )
...
Update pop2piano.md
2023-09-05 11:07:02 +01:00
Huazhong Ji
1cc3bc22fe
nn.Identity is not required to be compatible with PyTorch < 1.1.0 as the minimum PyTorch version we currently support is 1.10.0 ( #25974 )
...
nn.Identity is not required to be compatible with PyTorch < 1.1.0 as the
minimum PyTorch version we currently support is 1.10.0
2023-09-05 11:37:54 +02:00
Yih-Dar
fbbe1b8a40
Fix test_load_img_url_timeout ( #25976 )
...
* fix
* fix
* fix
---------
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
2023-09-05 11:34:28 +02:00
Yih-Dar
feec56959a
Fix Detr CI ( #25972 )
...
fix
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
2023-09-05 11:19:56 +02:00
Susnato Dhar
404ff8fc17
Fix typo ( #25966 )
...
* Update feature_extraction_clap.py
* changed all lenght to length
2023-09-05 10:12:25 +02:00
Lysandre
d8e13b3e04
v4.34.dev.0
2023-09-04 15:12:11 -04:00
Younes Belkada
49b69fe0d4
[Falcon] Remove SDPA for falcon to support earlier versions of PyTorch (< 2.0) ( #25947 )
...
* remove SDPA for falcon
* revert previous behaviour and add warning
* nit
* Update src/transformers/models/falcon/modeling_falcon.py
Co-authored-by: Matt <Rocketknight1@users.noreply.github.com>
* Update src/transformers/models/falcon/modeling_falcon.py
---------
Co-authored-by: Matt <Rocketknight1@users.noreply.github.com>
Co-authored-by: Lysandre Debut <hi@lysand.re>
2023-09-04 14:34:04 -04:00
Lysandre Debut
22a69f1d7d
Put Falcon back ( #25960 )
...
* Put Falcon back
* Update src/transformers/models/auto/configuration_auto.py
Co-authored-by: Arthur <48595927+ArthurZucker@users.noreply.github.com>
* Update test
---------
Co-authored-by: Arthur <48595927+ArthurZucker@users.noreply.github.com>
2023-09-04 14:17:09 -04:00
David Reguera
040c4613c2
Add type hints for tf models final batch ( #25883 )
...
* Add missing type hints and consistency to `RegNet` models
* Add missing type hints and consistency to `TFSamModel`
* Add missing type hints to `TFSegformerDecodeHead`
* Add missing type hints and consistency to `TransfoXL` family models
* Add missing type hints and consistency to `TFWav2Vec2ForSequenceClassification`
* Add type hints to `TFXLMModel`
* Fix linter
* Revert the type hints for `RegNet` to python 3.8 compliant
* Remove the redundant np.ndarray type hint.
2023-09-04 18:16:10 +01:00
Yih-Dar
44d2c199f6
Fix smart check ( #25955 )
...
* fix
* fix
* fix
---------
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
2023-09-04 18:54:34 +02:00
Lysandre Debut
3a479672ea
Fix failing test ( #25963 )
2023-09-04 12:53:50 -04:00
Matt
034bc5d26a
Add proper Falcon docs and conversion script ( #25954 )
...
* Add proper Falcon docs and conversion script
* Autodetect the decoder architecture instead of using an arg
* Update docs now that we can autodetect
* Fix doc error
* Add doc to toctree
* Quick doc update
2023-09-04 17:18:34 +01:00
Sanchit Gandhi
d750eff627
[VITS] Fix init test ( #25945 )
...
* [VITS] Fix init test
* add flaky decorator
* style
* max attempts
Co-authored-by: Yih-Dar <2521628+ydshieh@users.noreply.github.com>
* style
---------
Co-authored-by: Yih-Dar <2521628+ydshieh@users.noreply.github.com>
2023-09-04 17:09:26 +01:00
Nino Risteski
7cd01d4e38
Update README.md ( #25922 )
...
fixed a typo
2023-09-04 16:11:00 +02:00
Omar Sanseviero
bfb1895e33
Import deepspeed utilities from integrations ( #25919 )
...
Follow up from #25599
2023-09-04 14:03:48 +01:00
Sanchit Gandhi
eb984418e2
[VITS] Handle deprecated weight norm ( #25946 )
2023-09-04 11:54:03 +01:00
Sanchit Gandhi
f435003e0c
[MMS] Fix pip install in docs ( #25949 )
2023-09-04 11:53:41 +01:00
Nino Risteski
604a6c51ae
Update README.md ( #25941 )
...
fixed a typo
2023-09-04 11:28:21 +01:00
Nino Risteski
d4407a3bd1
Update autoclass_tutorial.md ( #25929 )
...
fixed typos
2023-09-04 11:16:49 +01:00
Nino Risteski
51e1e8120b
Update community.md ( #25928 )
...
fixed a few typos
2023-09-04 11:16:34 +01:00
omahs
0f0e1a2c2b
Fix typos ( #25936 )
...
* fix typo
* fix typo
* fix typo
* fix typos
* fix typos
* fix typo
* fix typo
* fix typo
* fix typos
* fix typo
* fix typo
* fix typo
* fix typos
* fix typos
2023-09-04 11:15:12 +01:00
Yih-Dar
b1d475f6d2
Skip offload tests for ViTDet ( #25913 )
...
* update
* fix
---------
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
2023-09-04 11:35:39 +02:00
ydshieh
ab8cba824e
CI: hotfix (skip VitsModelTest::test_initialization)
2023-09-04 09:06:11 +02:00
Nino Risteski
0afa5071bd
Update model_memory_anatomy.md ( #25896 )
...
typo fixes
2023-09-01 12:27:01 -07:00
Arthur
a4dd53d88e
Update-llama-code ( #25826 )
...
* some bug fixes
* updates
* Update code_llama.md
Co-authored-by: Omar Sanseviero <osanseviero@users.noreply.github.com>
* Add co author
Co-authored-by: pcuenca <pedro@latenitesoft.com>
* add a test
* fixup
* nits
* some updates
* fix-coies
* adress comments
* nits
* nits
* fix docsting
* Apply suggestions from code review
Co-authored-by: amyeroberts <22614925+amyeroberts@users.noreply.github.com>
* update
* add int for https://huggingface.co/spaces/hf-accelerate/model-memory-usage
---------
Co-authored-by: Omar Sanseviero <osanseviero@users.noreply.github.com>
Co-authored-by: pcuenca <pedro@latenitesoft.com>
Co-authored-by: amyeroberts <22614925+amyeroberts@users.noreply.github.com>
2023-09-01 20:40:40 +02:00
Sanchit Gandhi
3587769c08
[VITS] Only trigger tokenizer warning for uroman ( #25915 )
2023-09-01 19:27:01 +01:00
Sanchit Gandhi
1fa2d89a9b
[MMS] Update docs with HF TTS implementation ( #25907 )
...
* [MMS] Update docs with HF TTS implementation
* Update docs/source/en/model_doc/mms.md
Co-authored-by: Arthur <48595927+ArthurZucker@users.noreply.github.com>
* add uromanise to docs
---------
Co-authored-by: Arthur <48595927+ArthurZucker@users.noreply.github.com>
2023-09-01 16:50:59 +01:00
Sanchit Gandhi
b439129e74
[VITS] Add to TTA pipeline ( #25906 )
...
* [VITS] Add to TTA pipeline
* Update tests/pipelines/test_pipelines_text_to_audio.py
Co-authored-by: Yoach Lacombe <52246514+ylacombe@users.noreply.github.com>
* remove extra spaces
---------
Co-authored-by: Yoach Lacombe <52246514+ylacombe@users.noreply.github.com>
2023-09-01 16:39:00 +01:00
Zach Mueller
be0e189bd3
Revert frozen training arguments ( #25903 )
...
* Revert frozen training arguments
* TODO
2023-09-01 11:24:12 -04:00
Omar Sanseviero
69c5b8f186
Remove broken docs for MusicGen ( #25905 )
...
Update musicgen.md
2023-09-01 15:26:42 +01:00
Yih-Dar
16d6e3087c
Better error message for pipeline loading ( #25912 )
...
* update
* update
* update
* update
---------
Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
2023-09-01 16:09:12 +02:00
Joao Gante
53e2fd785b
Falcon: Add RoPE scaling ( #25878 )
2023-09-01 12:05:53 +01:00
pkumc
024acd271b
fix FSDP model resume optimizer & scheduler ( #25852 )
...
* fix FSDP resume optimizer & scheduler
* improve trainer code quality
---------
Co-authored-by: machi04 <machi04@meituan.com>
2023-09-01 15:20:42 +05:30
Matthijs Hollemans
4ece3b9433
add VITS model ( #24085 )
...
* add VITS model
* let's vits
* finish TextEncoder (mostly)
* rename VITS to Vits
* add StochasticDurationPredictor
* ads flow model
* add generator
* correctly set vocab size
* add tokenizer
* remove processor & feature extractor
* add PosteriorEncoder
* add missing weights to SDP
* also convert LJSpeech and VCTK checkpoints
* add training stuff in forward
* add placeholder tests for tokenizer
* add placeholder tests for model
* starting cleanup
* let the great renaming begin!
* use config
* global_conditioning
* more cleaning
* renaming variables
* more renaming
* more renaming
* it never ends
* reticulating the splines
* more renaming
* HiFi-GAN
* doc strings for main model
* fixup
* fix-copies
* don't make it a PreTrainedModel
* fixup
* rename config options
* remove training logic from forward pass
* simplify relative position
* use actual checkpoint
* style
* PR review fixes
* more review changes
* fixup
* more unit tests
* fixup
* fix doc test
* add integration test
* improve tokenizer tests
* add tokenizer integration test
* fix tests on GPU (gave OOM)
* conversion script can handle repos from hub
* add conversion script for all MMS-TTS checkpoints
* automatically create a README for the converted checkpoint
* small changes to config
* push README to hub
* only show uroman note for checkpoints that need it
* remove conversion script because code formatting breaks the readme
* make WaveNet layers configurable
* rename variables
* simplifying the math
* output attentions and hidden states
* remove VitsFlip in flow model
* also got rid of the other flip
* fix tests
* rename more variables
* rename tokenizer, add phonemization
* raise error when phonemizer missing
* re-order config docstrings to match method
* change config naming
* remove redundant str -> list
* fix copyright: vits authors -> kakao enterprise
* (mean, log_variances) -> (prior_mean, prior_log_variances)
* if return dict -> if not return dict
* speed -> speaking rate
* Apply suggestions from code review
Co-authored-by: amyeroberts <22614925+amyeroberts@users.noreply.github.com>
* update fused tanh sigmoid
* reduce dims in tester
* audio -> output_values
* audio -> output_values in tuple out
* fix return type
* fix return type
* make _unconstrained_rational_quadratic_spline a function
* all nn's to accept a config
* add spectro to output
* move {speaking rate, noise scale, noise scale duration} to config
* path -> attn_path
* idxs -> valid idxs -> padded idxs
* output values -> waveform
* use config for attention
* make generation work
* harden integration test
* add spectrogram to dict output
* tokenizer refactor
* make style
* remove 'fake' padding token
* harden tokenizer tests
* ron norm test
* fprop / save tests deterministic
* move uroman to tokenizer as much as possible
* better logger message
* fix vivit imports
* add uroman integration test
* make style
* up
* matthijs -> sanchit-gandhi
* fix tokenizer test
* make fix-copies
* fix dict comprehension
* fix config tests
* fix model tests
* make outputs consistent with reverse/not reverse
* fix key concat
* more model details
* add author
* return dict
* speaker error
* labels error
* Apply suggestions from code review
Co-authored-by: amyeroberts <22614925+amyeroberts@users.noreply.github.com>
* Update src/transformers/models/vits/convert_original_checkpoint.py
Co-authored-by: amyeroberts <22614925+amyeroberts@users.noreply.github.com>
* remove uromanize
* add docstrings
* add docstrings for tokenizer
* upper-case skip messages
* fix return dict
* style
* finish tests
* update checkpoints
* make style
* remove doctest file
* revert
* fix docstring
* fix tokenizer
* remove uroman integration test
* add sampling rate
* fix docs / docstrings
* style
* add sr to model output
* fix outputs
* style / copies
* fix docstring
* fix copies
* remove sr from model outputs
* Update utils/documentation_tests.txt
Co-authored-by: amyeroberts <22614925+amyeroberts@users.noreply.github.com>
* add sr as allowed attr
---------
Co-authored-by: sanchit-gandhi <sanchit@huggingface.co>
Co-authored-by: Sanchit Gandhi <93869735+sanchit-gandhi@users.noreply.github.com>
Co-authored-by: amyeroberts <22614925+amyeroberts@users.noreply.github.com>
2023-09-01 10:50:06 +01:00
