huggingface/transformers

Pull Requests Commits

Config kwargs (#44953)

zucchini-nlp committed 217 days ago

Verified d4895f08

Fix: Update optimization.py (#44909)

anshuS1310 committed 217 days ago

Verified 2f121eaf

[CB] [Minor] Simplify test suite (#44858)

remi-or committed 217 days ago

Verified 4f8de834

Allow arbitrary template kwargs in processors (#44881)

zucchini-nlp committed 217 days ago

Verified 9ef15d70

Fix missing post_processor in DebertaV2Tokenizer causing no special t… (#44570)

umbilnm committed 217 days ago

Verified 6a056a16

incorrect model list update (#44880)

itazap committed 217 days ago

Verified bc8c80e0

refactor: mlinter as its own package (#44939)

tarekziade committed 217 days ago

Verified a48a63c2

[CB] Add an option to return logprobs (#44835)

remi-or committed 218 days ago

Verified 19bba39f

[docs] peft (#44804)

stevhliu committed 218 days ago

Verified a9532bcf

Continuous batching thread safety (#44924)

Qubitium committed 218 days ago

Verified dda54684

Add static FP8 expert support (#44895)

SunMarc committed 218 days ago

Verified e5ad3946

Fix variable shadowing in pipeline example and typo in BART docs (BERT → BART) (#44935)

VanshikaSohal committed 218 days ago

Verified f6195948

Fix failing job `Update Transformers metadata` after #43514 (#44941)

ydshieh committed 218 days ago

Verified 0f19dec7

Clearer type hints and fix rope validation in configs (#44943)

zucchini-nlp committed 218 days ago

Verified ed003b44

Correct docstrings for `from_pretrained` (url input deprecated) (#44946)

BSchilperoort committed 218 days ago

Verified eb3d67aa

fix(i18n): replace broken relative links to awesome-transformers.md with absolute URLs (#44905)

NicoleRobin committed 218 days ago

Verified ddb709c8

Fix backward compatibility for full path imports of Fast Image Processors (#44926)

yonigozlan committed 218 days ago

Verified d03945fb

chore(typing): added rule 11 (#44865)

tarekziade committed 218 days ago

Verified 29db503c

fix: improve processor loading performance by avoiding redundant tokenizer parsing (#44927)

ydshieh committed 218 days ago

Verified b1527a32

fix(camembert): add tie_word_embeddings=True to CamembertConfig (#44931)

r266-tech committed 218 days ago

Verified 9dc8d8aa

Support SizeDict import in get_size_dict (#44903)

yonigozlan committed 218 days ago

Verified 687a70da

fix `processing_utils.py`: avoid deepcopying tokenizer in `ProcessorMixin` to improve performance (#44894)

ydshieh committed 218 days ago

Verified a8683756

fix: set `clean_up_tokenization_spaces=False` in Llama 3 tokenizer conversion (#44914)

maxsloef-goodfire committed 218 days ago

Verified 55cc1a7f

Add big angry code agent warnings! (#44890)

Rocketknight1 committed 218 days ago

Verified 394b1a0e

[docs] model cards (#44837)

stevhliu committed 220 days ago

Verified 3a3b59cb

[Model] Add UVDoc Model Support (#43385)

XingweiDeng committed 220 days ago

Verified 52bc9b7e

Add backward compatibility for direct imports from legacy `image_processing_utils_fast` (#44897)

yonigozlan committed 220 days ago

Verified e168f86e

[`FA4`] Add kernels fallback (#44797)

vasqu committed 220 days ago

Verified e6ed96c7

Bump kernels version dependency to avoid crashes (#44887)

Cyrilvallez committed 220 days ago

Verified a269c990

[Model] Add SLANeXt Model Support (#43707)

liu-jiaxuan committed 221 days ago

Verified d4f88c20

Newer Older