huggingface/transformers

Pull Requests Commits

drop_py38_build_img_2

ydshieh committed 1 year ago

25db886d

update docker files

ydshieh committed 1 year ago

f1400287

drop python 3.8

ydshieh committed 1 year ago

175eb4cd

Add option for running ffmpeg_microphone_live as a background process (#32838)

mikamerath committed 1 year ago

Verified eef6b0ba

Olmo is ExecuTorch Compatible (#34181)

guangy10 committed 1 year ago

Verified c14ccbcd

Qwen2.5 is ExecuTorch Compatible (#34102)

guangy10 committed 1 year ago

Verified 7a08a772

Add post_process_depth_estimation to image processors and support ZoeDepth's inference intricacies (#32550)

alex-bene committed 1 year ago

Verified c31a6ff4

Fix: tensor of examples of the same length triggers invalid stacking (#34166)

pbelcak committed 1 year ago

Verified 104599d7

Fix FA2 attention for models supporting sliding window (#34093)

Cyrilvallez committed 1 year ago

Verified 51e395d1

[RT-DETR] Fix onnx inference bug for Optype (Where) (#33877)

YHallouard committed 1 year ago

Verified eb6a7349

Update PR templates (#34065)

SunMarc committed 1 year ago

Verified 84b17e03

Sync video classification pipeline with huggingface_hub spec (#34288)

Rocketknight1 committed 1 year ago

Verified 681fc437

Fix Korean doc _toctree.yml (#34293)

regisss committed 1 year ago

Verified 93352e81

[docs] Fix GenerationConfig params (#34299)

stevhliu committed 1 year ago

Verified b644178e

T5 compile compatibilty (#34089)

zucchini-nlp committed 1 year ago

Verified 73d65e63

VLM: add more modularity (#34175)

zucchini-nlp committed 1 year ago

Verified 5077bc03

Attn implementation for composite models (#32238)

zucchini-nlp committed 1 year ago

Verified 21d50258

Fix method name which changes in tutorial (#34252)

andimarafioti committed 1 year ago

Verified 32590b5e

Add a doc section on writing generation prompts (#34248)

Rocketknight1 committed 1 year ago

Verified f701b98e

Add DetrImageProcessorFast (#34063)

yonigozlan committed 1 year ago

Verified a4122813

Change Paligemma import logging to work with modular (#34211)

yonigozlan committed 1 year ago

Verified 24bdc94d

Generation tests: don't rely on main input name (#34228)

zucchini-nlp committed 1 year ago

Verified ca541bd4

Only cast logits to float when computing loss (#34147)

Matthew Hoffman committed 1 year ago

Verified 816f4424

Fix UDOP dtype issue (#34180)

Rocketknight1 committed 1 year ago

Verified e46e3bc1

add Glm (#33823)

Cyrilvallez committed 1 year ago

Verified 66047640

Informative 2 (#34154)

LysandreJik committed 1 year ago

Verified e95ea479

Fix broken test decorator `require_torch_up_to_2_accelerators` (#34201)

byi8220 committed 1 year ago

Verified 0437d6cd

BLIP: fix input expansion logic (#34225)

zucchini-nlp committed 1 year ago

Verified 5a5b590d

Fix-red-ci (#34230)

ArthurZucker committed 1 year ago

Verified b54109c7

Enable users to use their own loss functions + deal with prefetching for grad accum (#34198)

muellerzr committed 1 year ago

Verified 6ba31a8a

Older