huggingface/transformers

Pull Requests Commits

Universal Speculative Decoding `CandidateGenerator` (#35029)

keyboardAnt committed 347 days ago

Verified d18d9c32

fix: prevent model access error during Optuna hyperparameter tuning (#36395)

emapco committed 347 days ago

Verified 082834dd

add recommendations for NPU using flash_attn (#36383)

zheliuyu committed 347 days ago

Verified 6513e5e4

Fixing the docs corresponding to the breaking change in torch 2.6. (#36420)

Narsil committed 347 days ago

Verified b4965cec

Deprecate transformers.agents (#36415)

aymeric-roucher committed 347 days ago

Verified 9a217fc3

Add retry hf hub decorator (#35213)

muellerzr committed 348 days ago

Verified 41925e42

Fixed VitDet for non-squre Images (#35969)

cjfghk5697 committed 348 days ago

Verified 9ebfda32

Security fix for `benchmark.yml` (#36402)

ydshieh committed 348 days ago

Verified cbe0ea59

Fix convert_to_rgb for SAM ImageProcessor (#36369)

MSt-10 committed 348 days ago

Verified 88d10517

[CLI] add import guards (#36376)

gante committed 348 days ago

Verified e1ce9489

Fix pytorch integration tests for SAM (#36397)

qubvel committed 348 days ago

Verified fb83befb

chore: fix function argument descriptions (#36392)

threewebcode committed 348 days ago

Verified ca6ebcb9

fix audio classification pipeline fp16 test on cuda (#36359)

jiqing-feng committed 348 days ago

Verified 7c8916dd

[tests] enable autoawq tests on XPU (#36327)

faaany committed 348 days ago

Verified c3700b0e

tests: revert change of torch_require_multi_gpu to be device agnostic (#35721)

dvrogozh committed 348 days ago

Verified b4b9da6d

addressing the issue #34611 to make FlaxDinov2 compatible with any batch size (#35138)

MHRDYN7 committed 348 days ago

Verified d80d52b0

Added handling for length <2 of suppress_tokens for whisper (#36336)

andreystarenky committed 348 days ago

Verified 3a02fe56

Fix doc formatting in forward passes & modular (#36243)

Cyrilvallez committed 348 days ago

Verified da4ab2a1

Update _get_eval_sampler to reflect Trainer.tokenizer is deprecation self.tokenizer -> self.processing_class (#36315)

yukiman76 committed 348 days ago

Verified 92abc0da

enable torchao quantization on CPU (#36146)

jiqing-feng committed 348 days ago

Verified 9d6abf97

Fix `is_causal` fail with compile (#36374)

Cyrilvallez committed 348 days ago

Verified 401543a8

[modular] Do not track imports in functions (#36279)

Cyrilvallez committed 348 days ago

Verified bc65f3fc

Load models much faster on accelerator devices!! (#36380)

Cyrilvallez committed 348 days ago

Verified 4b5cf549

Update modeling_llava_onevision.py (#36391)

yinsong1986 committed 348 days ago

Verified 931e5f4a

notify new model merged to `main` (#36375)

ydshieh committed 349 days ago

Verified 2ab7bdc4

[Modeling] Reduce runtime when loading missing keys (#36312)

kylesayrs committed 349 days ago

Verified 05dfed06

fix(type): padding_side type should be Optional[str] (#36326)

shenxiangzhuang committed 349 days ago

Verified 18276b03

Update amd pytorch index to match base image (#36347)

ivarflakstad committed 349 days ago

Verified f4684a6e

Add autoquant support for torchao quantizer (#35503)

jerryzh168 committed 349 days ago

Verified 2af272c1

Change slack channel for mi250 CI to amd-hf-ci (#36346)

ivarflakstad committed 349 days ago

Verified 977a61f7

Newer Older