huggingface/transformers

Pull Requests Commits

Remove 'main' from doc links (#22860)

amyeroberts committed 3 years ago

Verified 5f97bbc1

use `accelerate@main` in CI (#22859)

ydshieh committed 3 years ago

Verified 4603fe9b

feat(model parallelism): move labels to the same device as logits for M2M100 (#22850)

elabongaatuo committed 3 years ago

Verified 337225ec

move preprocess_logits_for_metrics before _nested_gather in trainer.e… (#22603)

ChenyangLiu committed 3 years ago

Verified 6bd8ae26

fix SpeechT5 doc comments (#22854)

hollance committed 3 years ago

Verified c582e8aa

Make ClipSeg compatible with model parallelism (#22844)

youssefadr committed 3 years ago

Verified 84a6570e

Raise err if minimum Accelerate version isn't available (#22841)

muellerzr committed 3 years ago

Verified 5bb4ec62

Fix from_pretrained when model is instantiated on the meta device (#22837)

sgugger committed 3 years ago

Verified 5f092194

Use code on the Hub from another repo (#22814)

sgugger committed 3 years ago

Verified 5f9b825c

Update accelerate version + warning check fix (#22833)

muellerzr committed 3 years ago

Verified aec10d16

Generate: Add assisted generation (#22211)

gante committed 3 years ago

Verified 78cda46f

Fix `test_eos_token_id_int_and_list_top_k_top_sampling` (#22826)

ydshieh committed 3 years ago

Verified 90247d3e

Fix Past CI not running against the latest `main` (#22823)

ydshieh committed 3 years ago

Verified 1ebc1dee

🌐 [i18n-KO] Fix anchor links for docs `auto_tutorial`, `training` (#22796)

gabrielwithappy committed 3 years ago

Verified 42288269

TTS fine-tuning for SpeechT5 (#21824)

hollance committed 3 years ago

Verified ac2bc50a

Mark auto models as important (#22815)

sgugger committed 3 years ago

Verified dacd3456

Introduce `PartialState` as the device handler in the `Trainer` (#22752)

muellerzr committed 3 years ago

Verified 03462875

Revert "Use code on the Hub from another repo" (#22813)

sgugger committed 3 years ago

Verified 50caa206

Simplify update metadata job (#22811)

sgugger committed 3 years ago

Verified e13d6ef7

Remove accelerate from tf test reqs (#22777)

muellerzr committed 3 years ago

Verified cd3e0211

Fix squeeze into torch 1.x compatible form in llama model (#22808)

DyeKuu committed 3 years ago

Verified f8c43c94

Don't use `LayoutLMv2` and `LayoutLMv3` in some pipeline tests (#22774)

ydshieh committed 3 years ago

Verified 5269718c

Use code on the Hub from another repo (#22698)

sgugger committed 3 years ago

Verified ea7b0a53

🌐 [i18n-KO] Translated `tasks/translation.mdx` to Korean (#22805)

wonhyeongseo committed 3 years ago

Verified 4d2c52e8

Fix sneaky torch dependency in TF example (#22804)

Rocketknight1 committed 3 years ago

Verified 2237127a

improve(llama): Faster apply_rotary_pos_emb (#22785)

fpgaminer committed 3 years ago

Verified 626c1b8a

[i18n-KO] fix: docs: ko: sagemaker anchors and `_toctree.yml` (#22549)

jungnerd committed 3 years ago

Verified abbc96a2

🌐 [i18n-KO] Translated `custom_models.mdx` to Korean (#22534)

HanNayeoniee committed 3 years ago

Verified 18c89481

Fix `test_word_time_stamp_integration` for `Wav2Vec2ProcessorWithLMTest` (#22800)

ydshieh committed 3 years ago

Verified 76d24f1a

Generate: add CJK support to TextStreamer (#22664)

bcol23 committed 3 years ago

Verified 28f26c10

Newer Older