huggingface/transformers

Pull Requests Commits

fix the error of tests/quantization/fbgemm_fp8/test_fbgemm_fp8.py::Fb… (#43547)

sywangyi committed 3 days ago

Verified b33235e3

[Trainer] Move optimizer cls init to trainer_optimizer.py (#43738)

SunMarc committed 3 days ago

Verified 760eae30

Make it possible to evaluate when using sequence parallel in HF Trainer (#43517)

jp1924 committed 3 days ago

Verified af496d74

Remove unnecessary code or checks for PT 2.4+ (#43787)

cyyever committed 3 days ago

Verified 520fad98

Update TrainingArguments (#43806)

SunMarc committed 3 days ago

Verified 44f92b63

Fix old tech stack in doc (#43879)

cyyever committed 3 days ago

Verified 6028dd2b

Fix SwanLab callback to forward resume init args (#43848)

OiPunk committed 3 days ago

Verified 6df88fab

[`Jamba`] Fallback to slow path and warn instead of error out (#43889)

vasqu committed 3 days ago

Verified 9b4de431

[MistralCommonBackend] fix loading proc (#43887)

eustlb committed 3 days ago

Verified 3b2abb8f

Fix GlmMoeDsaConfig default mlp_layer_types in modular conversion (#43876)

OiPunk committed 3 days ago

Verified 476600a9

Fix init weights in remote code (#43768)

zucchini-nlp committed 3 days ago

Verified 884749a1

Improve PR comment CI feedback (#43852)

ydshieh committed 3 days ago

Verified b7b9d252

docs: correct typo 'neccessary' to 'necessary' (#43868)

thecaptain789 committed 4 days ago

Verified b2028e77

[xLSTM] Fix bugs preventing small model training (#43209)

Anri-Lombard committed 4 days ago

Verified a47a89a4

docs: Add Switch Transformers docstring notes and update spectrogram comment (#43336)

harshaljanjani committed 4 days ago

Verified 82a5b3d7

Fix cardinality error for DETR models without explicit background class (#43513)

heathdutton committed 4 days ago

Verified d7ac5cd0

Look for the pad_token_id in the right place for Llama4 (#43539)

Rocketknight1 committed 4 days ago

Verified 9e4a8c44

:rotating_light: [`Attn`] New attn mask interface everywhere (#42848)

vasqu committed 4 days ago

Verified 4b8ba25a

Fix video interpolation in pe_audio_video (#43811)

Rocketknight1 committed 4 days ago

Verified 7d5681b3

[docs] Remove pipeline() examples from summarization/translation tasks (#43831)

Mr-Neutr0n committed 4 days ago

Verified 018af00b

Update ModelType for Unigram tokenizer (#43860)

pavel-esir committed 4 days ago

Verified 1f62d4ed

Add GlmMoeDsa (#43858)

Cyrilvallez committed 4 days ago

Verified 33a8e68d

delete unnecessary code to make moe compatible to full graph compile (#43855)

kaixuanliu committed 4 days ago

Verified 3b4a064d

Delete batch_split from EncoderDecoderCache (#43814)

cyyever committed 4 days ago

Verified 0f4681a3

Adding Support for Qwen3.5 (#43830)

bozheng-hit committed 4 days ago

Verified fc913722

fix gptoss crash in tp (#43853)

sywangyi committed 4 days ago

Verified 161c4303

Add id and resume support to SwanLab integration (#43719)

i-pj committed 4 days ago

Verified 41833e3e

Remove GPU tracking from TrackioCallback and remove env var support (#43371)

qgallouedec committed 4 days ago

Verified ab830e29

Fix incorrect timestamp calculation in Qwen3VL Processor (#43659)

jonathan-fulton committed 4 days ago

Verified 2c3926ec

add `do_sample=False` to qwen2_5_vl model tests to stablize the output (#43728)

kaixuanliu committed 4 days ago

Verified e0cb4d18

Newer Older