huggingface/transformers

Pull Requests Commits

ydshieh committed 346 days ago

d68cbd5e

ydshieh committed 346 days ago

500519c8

ydshieh committed 346 days ago

748beb83

ydshieh committed 346 days ago

bc5fa46a

ydshieh committed 346 days ago

3cae9734

fix: Fully remove legacy cache from Llama (#36958)

Wheest committed 347 days ago

Verified 348f3285

fixed typo (#37036)

finnoh committed 347 days ago

Verified d6b3c748

Remove deprecated batch_size parameter (#37007)

cyyever committed 347 days ago

Verified 6cc9c8d7

Replace default split function with jnp.split() in flax models (#37001)

premmurugan229 committed 347 days ago

Verified 4cc65e99

Set weights_only in torch.load (#36991)

cyyever committed 347 days ago

Verified 41a0e58e

Fix typing for None valued variables (#37004)

cyyever committed 347 days ago

Verified de77f5b1

Avoid unnecessary device operations in loss computing (#36950)

cyyever committed 347 days ago

Verified 8c5e29ba

clean pipeline question_answering. (#36986)

zhanluxianshen committed 347 days ago

Verified 471cf1de

[generate, cache] handle more complex device maps (#37014)

gante committed 347 days ago

Verified 29f322d0

[audio utils] fix fft_bin_width computation (#36603)

eustlb committed 347 days ago

Verified fb8e6c50

[chat templates} support loading audio from video (#36955)

zucchini-nlp committed 347 days ago

Verified e97c7600

Fixup for distill_any_depth conversion script (#37043)

qubvel committed 347 days ago

Verified c7bc79bd

Optimize `to_py_obj` for python-native numeric lists and scalars (#36885)

n0gu-furiosa committed 347 days ago

Verified d1eafe8d

fix pegasus init weights and other copied models (#36844)

jiqing-feng committed 347 days ago

Verified 0e56fb69

Add Distill Any Depth (#36614)

keetrap committed 347 days ago

Verified 7e813f9c

Skip FP8 linear tests For device capability < 9.0(#37008)

MekkCyber committed 347 days ago

Verified 92429057

remove redundant code in trainer (#36994)

hiyouga committed 347 days ago

Verified 279c2e30

Mark 2 tests as flaky for now (#37038)

ydshieh committed 347 days ago

Verified d13c390d

[Modeling] Load FP8 safetensors such as DeepSeek (#36828)

kylesayrs committed 347 days ago

Verified d6d930a6

Fix PixtralProcessor patch_size when spatial_merge_size is used (#37019)

mgoin committed 347 days ago

Verified 927ce1d3

Support QuestionAnswering Module for ModernBert based models. (#35566)

bakrianoo committed 348 days ago

Verified 49b5ab6a

fix transformers_cli import relative path issue (#36989)

yao-matrix committed 348 days ago

Verified 5b08db88

[docs] Attention mask image (#36970)

stevhliu committed 348 days ago

Verified 3a8ec8c4

Remove deprecated training arguments (#36946)

cyyever committed 348 days ago

Verified 2b550c47

fix typos in the code comments and error messages (#36993)

threewebcode committed 348 days ago

Verified 44715225

Older