huggingface/transformers

Pull Requests Commits

Fix DocQA max_answer_len validation error message (#42948)

antznette1 committed 105 days ago

Verified 64ec2ba1

Fix formatting of trackio model tag (#42973)

abidlabs committed 105 days ago

Verified 817886a6

Hardcode the factor in caching allocator (#42996)

Cyrilvallez committed 105 days ago

Verified f5d9d808

Do not use global variable, and improve context manager coverage (#42957)

Cyrilvallez committed 105 days ago

Verified 0d2dbaa9

[`Tests`] Fix inputs placement (#42963)

vasqu committed 105 days ago

Verified 007274db

fix device dismatch issue for pe_audio_video model parallelism (#42917)

kaixuanliu committed 105 days ago

Verified 1b280f99

Single config attribute for weight tying (#42815)

zucchini-nlp committed 105 days ago

Verified 9a90500b

Fix tests trainer again (#42933)

SunMarc committed 108 days ago

Verified 70179949

[`Ernie 4.5`] Ernie VL models (#39585)

vasqu committed 108 days ago

Verified a8a22624

[`Generation`] Fix default overwrite for non-`None` defaults (#42958)

vasqu committed 108 days ago

Verified f218ed21

Fix infinity in JSON serialized files (#42959)

LysandreJik committed 108 days ago

Verified d14d99ea

Add runner specification to CodeQL workflow (#42955)

paulinebm committed 108 days ago

Verified d54d78f7

[loading] Really initialize on meta device for huge perf gains (#42941)

Cyrilvallez committed 108 days ago

Verified bb9357ff

fix error: 'BlockMask' object has no attribute 'dtype' for lasr model (#42802)

kaixuanliu committed 108 days ago

Verified 789226c1

fix concat order (#42946)

eustlb committed 108 days ago

Verified a4d62291

[kernels] Fix failling tests (#42953)

MekkCyber committed 108 days ago

Verified cf0f071e

[Quantization] CI green by end of year (#42951)

MekkCyber committed 108 days ago

Verified b5eea347

Updated `backbone_config` docstrings and type annotations (#42927)

simonreise committed 109 days ago

Verified 3e4baf8e

[docs] dtype (#42883)

stevhliu committed 109 days ago

Verified b93f2e3a

Load generation config from nested configs (#42922)

zucchini-nlp committed 109 days ago

Verified f2c6d2ad

🚨 Fix ConvNeXt image processor default interpolation to BICUBIC (#42934)

lukepayyapilli committed 109 days ago

Verified 60634caa

fix(tvp): add missing type_vocab_size parameter to TvpConfig (#42928)

majiayu000 committed 109 days ago

Verified b62e5b3e

[Quantization] Removing misleading int8 quantization in Finegrained FP8 (#42945)

MekkCyber committed 109 days ago

Verified 4dc60fd3

Fp8 dq (#42926)

SunMarc committed 109 days ago

Verified af91c0ba

Add buffers to `_init_weights` for ALL models (#42309)

Cyrilvallez committed 109 days ago

Verified 537c2e3d

rewrite _process_parameter_type in auto_docstring.py to improve usability (#42431)

Yacklin committed 109 days ago

Verified 728f34c3

Update `param_element_size` (#42818)

SunMarc committed 109 days ago

Verified dd8057af

[Tests] Fix CompressedTensors tests (#42935)

kylesayrs committed 109 days ago

Verified 0a846542

[CB] Allow block sharing in hybrid models (#42877)

remi-or committed 109 days ago

Verified 04e78e67

[Quantization] Misc tests fixes (#42940)

MekkCyber committed 109 days ago

Verified d7dd443a

Newer Older