huggingface/transformers

Pull Requests Commits

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

zucchini-nlp committed 13 days ago

Verified ed003b44

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

BSchilperoort committed 13 days ago

Verified eb3d67aa

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

NicoleRobin committed 13 days ago

Verified ddb709c8

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

yonigozlan committed 13 days ago

Verified d03945fb

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

tarekziade committed 13 days ago

Verified 29db503c

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

ydshieh committed 13 days ago

Verified b1527a32

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

r266-tech committed 13 days ago

Verified 9dc8d8aa

Support SizeDict import in get_size_dict (#44903)

yonigozlan committed 13 days ago

Verified 687a70da

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

ydshieh committed 13 days ago

Verified a8683756

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

maxsloef-goodfire committed 13 days ago

Verified 55cc1a7f

Add big angry code agent warnings! (#44890)

Rocketknight1 committed 14 days ago

Verified 394b1a0e

[docs] model cards (#44837)

stevhliu committed 15 days ago

Verified 3a3b59cb

[Model] Add UVDoc Model Support (#43385)

XingweiDeng committed 15 days ago

Verified 52bc9b7e

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

yonigozlan committed 16 days ago

Verified e168f86e

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

vasqu committed 16 days ago

Verified e6ed96c7

Bump kernels version dependency to avoid crashes (#44887)

Cyrilvallez committed 16 days ago

Verified a269c990

[Model] Add SLANeXt Model Support (#43707)

liu-jiaxuan committed 16 days ago

Verified d4f88c20

Fix core dumped when `NemotronH` is torch compiled (#44854)

ydshieh committed 16 days ago

Verified e8712550

Fix several based models' pipeline parallel support (#44699)

hmellor committed 16 days ago

Verified 20a233bd

fix(testing): Fix PaliGemma 2 and PaddleOCR-VL test failures on main (#44765)

harshaljanjani committed 16 days ago

Verified 81128509

Fix dtype guessing from state dict (#44883)

Cyrilvallez committed 16 days ago

Verified 82db888e

Add missing dunder methods to `SizeDict` (#44884)

hmellor committed 16 days ago

Verified 7cd9b985

Fix VL model rope_deltas batch size mismatch in online RL training (#44873)

sergiopaniego committed 16 days ago

Verified b7164eca

Fix `layer_types` type hint for `AFMoE` and `Llama4` (#44874)

hmellor committed 16 days ago

Verified 1229e90d

Align lfm2 cache to other mamba caches (#44866)

Cyrilvallez committed 16 days ago

Verified cef28300

Fix nemotron config docstrings (#44878)

Cyrilvallez committed 16 days ago

Verified f5e57308

Fix nemotron_h modular (#44876)

Cyrilvallez committed 16 days ago

Verified 8dc7a52d

feat: added cache to the model linter (#44790)

tarekziade committed 16 days ago

Verified 2cd52c26

[Model] Add PP-Chart2Table Model Support (#43767)

XingweiDeng committed 17 days ago

Verified aa1c36f1

[Mistral] Fix query scaling for Mistral4 and Ministral3 (#44860)

Cyrilvallez committed 17 days ago

Verified b96f8a98

Newer Older