huggingface/transformers

Pull Requests Commits

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

Support Llama 3.2 conversion (text models) (#33778)

pcuenca committed 1 year ago

Verified 7a06d07e

Fix Gradient Accumulation issue (#34191)

ArthurZucker committed 1 year ago

Verified c1c7e896

Generate: visit non-llm `prepare_inputs_for_generation` (#34199)

gante committed 1 year ago

Verified f51ac9e0

Fix bus error when using GPT2 on M1 macs (#34031)

chanind committed 1 year ago

Verified 1d2c29f0

Llama3 and Llama2 are ExecuTorch compatible (#34101)

guangy10 committed 1 year ago

Verified 9470c000

removes decord (#33987)

vrnvu committed 1 year ago

Verified 7f508850

Fix for tokenizer.apply_chat_template with continue_final_message=True (#34214)

schoennenbeck committed 1 year ago

Verified f2846ad2

fix(Wav2Vec2ForCTC): torch export (#34023)

chrsmcgrr committed 1 year ago

Verified b57c7bce

Ping team members for new failed tests in daily CI (#34171)

ydshieh committed 1 year ago

Verified fce1fcfe

Fix warning message for fp32_cpu_offloading in bitsandbytes configs (#34079)

amosyou committed 1 year ago

Verified aa3e35ac

Update `trainer._get_eval_sampler()` to support `group_by_length` arg (#33514)

larin92 committed 1 year ago

Verified 6d2b2033

Revert "Fix FSDP resume Initialization issue" (#34193)

SunMarc committed 1 year ago

Verified 3f06f95e

Newer Older