huggingface/transformers

Pull Requests Commits

pkooij committed 309 days ago

Verified dcddb970

add openpi changes

pkooij committed 309 days ago

Verified 3676fc80

kashif committed 1 year ago

a5923d4d

refactor: remove `set_tracer_provider` and `set_meter_provider` calls (#39422)

McPatate committed 1 year ago

d04a9425

Cyrilvallez committed 1 year ago

Verified 37f8b0b5

finally fix glm4v accordingly

Cyrilvallez committed 1 year ago

Verified d15e157d

fix attr post conflict

Cyrilvallez committed 1 year ago

Verified 9873b2f9

Cyrilvallez committed 1 year ago

Verified d6f31553

Add a default value for `position_ids` in masking_utils (#39310)

Cyrilvallez committed 1 year ago

Verified cbac4263

fix Glm4v batch videos forward (#39172)

Kuangdd01 committed 1 year ago

Verified c1221458

[sliding window] revert and deprecate (#39301)

zucchini-nlp committed 1 year ago

Verified 9da3f7d0

[smollm3] add tokenizer mapping for `smollm3` (#39271)

gante committed 1 year ago

Verified b2fbc022

[pagged-attention] fix off-by-1 error in pagged attention generation (#39258)

kashif committed 1 year ago

Verified c30a6b7b

Fix errors when use verl to train GLM4.1v model (#39199)

kaln27 committed 1 year ago

Verified be7d1a9d

[bugfix] fix flash attention 2 unavailable error on Ascend NPU (#39166)

FightingZhen committed 1 year ago

Verified 6023ca8a

Fix some bug for finetune and batch infer For GLM-4.1V (#39090)

zRzRzRzRzRzRzR committed 1 year ago

Verified e7e78b2a

Release: v4.53.2

Cyrilvallez committed 1 year ago

Verified 4a5d5c49

Release: v4.53.1

Cyrilvallez committed 1 year ago

Verified 896e9cea

Add packed tensor format support for flex/sdpa/eager through the mask! (#39194)

Cyrilvallez committed 1 year ago

Verified bc0a47e6

when delaying optimizer creation only prepare the model (#39152)

winglian committed 1 year ago

Verified 63af3d7a

Fix multimodal processor get duplicate arguments when receive kwargs for initialization (#39125)

Isotr0py committed 1 year ago

Verified a10c2bf8

[smolvlm] fix video inference (#39147)

zucchini-nlp committed 1 year ago

Verified 650bc679

[qwen2-vl] fix FA2 inference (#39121)

zucchini-nlp committed 1 year ago

Verified 3c5f9108

Several fixes for Gemma3n (#39135)

Cyrilvallez committed 1 year ago

Verified 5e1c9146

Fix key mapping for VLMs (#39029)

bvantuan committed 1 year ago

Verified 8446e2a5

Fix: unprotected import of tp plugin (#39083)

S1ro1 committed 1 year ago

Verified 6900fe8a

Release: v4.53.0

LysandreJik committed 1 year ago

67ddc82f

[Modeling] Fix encoder CPU offloading for whisper (#38994)

kylesayrs committed 1 year ago

Verified 0a8081b0

Gemma 3n (#39059)

RyanMullins committed 1 year ago

Verified c63cfd6a

[tests] remove tests from libraries with deprecated support (flax, tensorflow_text, ...) (#39051)

gante committed 1 year ago

Verified 3e5cc128

Older