huggingface/transformers

Pull Requests Commits

ydshieh committed 1 year ago

26c76151

Fix flaky `test_batching_equivalence` (#35564)

ydshieh committed 1 year ago

Verified 1b2f942a

Setup loss_type in config at model init time (#34616)

ChanderG committed 1 year ago

Verified 4adc415b

Re-add missing __all__ for Cohere and Phi3 (#35578)

Cyrilvallez committed 1 year ago

Verified c8ab6ce6

Minor fix in video text 2 text docs (#35546)

merveenoyan committed 1 year ago

Verified 487c31a2

More model refactoring! (#35359)

Cyrilvallez committed 1 year ago

Verified 965a2fb3

Don't show warning for `inv_freq` buffers (#35255)

zucchini-nlp committed 1 year ago

Verified 137965ca

Fix multi-gpu loss (#35395)

ArthurZucker committed 1 year ago

Verified 8cad65a6

update code owners (#35576)

ArthurZucker committed 1 year ago

Verified 2e2f8015

[i18n-ar] Translated file: `docs/source/ar/tasks/multiple_choice.md` into Arabic (#35199)

AhmedAlmaghz committed 1 year ago

Verified a6256ec0

Fix all output_dir in test_trainer.py to use tmp_dir (#35266)

nhamanasu committed 1 year ago

Verified b32938ae

Pipeline: simple API for assisted generation (#34504)

gante committed 1 year ago

Verified 76da6ca0

[`PixtralLarge`] Update Pixtral conversion script to support large format! (#34801)

ArthurZucker committed 1 year ago

Verified 3f483bea

[docs] Remove Hiera from AUDIO MODELS in docs (#35544)

ADAning committed 1 year ago

Verified 4c2c12b3

ovewrite top_k when crate audio classification pipeline (#35541)

HERIUN committed 1 year ago

Verified 854dc794

add code owners (#35528)

ArthurZucker committed 1 year ago

Verified 8c555ca3

Add ViTPose (#30530)

NielsRogge committed 1 year ago

Verified 8490d315

fix: Qwen2-VL generate with inputs_embeds (#35466)

minostauros committed 1 year ago

Verified 4349a0e4

Update doc for `metric_for_best_model` when `save_strategy="best"`. (#35389)

seanswyi committed 1 year ago

Verified 88e18b3c

Add: num_additional_image_tokens to models (#35052)

jp1924 committed 1 year ago

Verified 29e74b7c

Enable auto task for timm models in pipeline (#35531)

qubvel committed 1 year ago

Verified 657bb14f

Bump torch requirement to >= 2 (#35479)

oraluben committed 1 year ago

Verified 1a6c1d3a

Timm wrapper label names (#35553)

qubvel committed 1 year ago

Verified 59e5b3f0

Update missing model error message (#35370)

Rocketknight1 committed 1 year ago

Verified f1639ea5

Update doc and default value of TextNetImageProcessor (#35563)

jadechoghari committed 1 year ago

Verified bd39b062

Add support for modular with fast image processors (#35379)

yonigozlan committed 1 year ago

Verified 651cfb40

[Docs] links to `logits-processor-zoo` (#35552)

gante committed 1 year ago

Verified 430d3d43

Fix Qwen2VL processor to handle odd number of frames (#35431)

jla524 committed 1 year ago

Verified 3c1895aa

support chat generator as input of TextGenerationPipeline (#35551)

lhoestq committed 1 year ago

Verified 3fde88b1

Pass correct `num_items_in_batch` value into the training_step function (#35438)

hiyouga committed 1 year ago

Verified ebdd1ad4

Older