huggingface/transformers

Pull Requests Commits

[`quality`] update quality check to make sure we check imports 😈 (#29771)

ArthurZucker committed 1 year ago

Verified e68ff304

Change in-place operations to out-of-place in LogitsProcessors (#29680)

zucchini-nlp committed 1 year ago

Verified fadb0533

Prepend `bos token` to Blip generations (#29642)

zucchini-nlp committed 1 year ago

Verified b469ebc5

Llama: always convert the causal mask in the SDPA code path (#29663)

gante committed 1 year ago

Verified ee38fc31

Generate: remove legacy generation mixin imports (#29782)

gante committed 1 year ago

Verified 5ffef2a9

Add support for `torch_dtype` in the run_mlm example (#29776)

jla524 committed 1 year ago

Verified ef6e371d

Add deterministic config to `set_seed` (#29778)

muellerzr committed 1 year ago

Verified 10d232e8

Silence deprecations and use the DataLoaderConfig (#29779)

muellerzr committed 1 year ago

Verified f0bfb150

Cast bfloat16 to float32 for Numpy conversions (#29755)

Rocketknight1 committed 1 year ago

Verified de627f5a

[`LlavaNext`] Fix llava next unsafe imports (#29773)

ArthurZucker committed 1 year ago

Verified 73a73b41

Fix docker image build for `Latest PyTorch + TensorFlow [dev]` (#29764)

ydshieh committed 1 year ago

Verified 2ddceef9

fix issue with logit processor during beam search in Flax (#29636)

giganttheo committed 1 year ago

Verified fd734be1

Allow `-OO` mode for `docstring_decorator` (#29689)

matthid committed 1 year ago

Verified 691c3d73

OWL-ViT box_predictor inefficiency issue (#29712)

RVV-karma committed 1 year ago

Verified 9556054f

Fixed typo in quantization_config.py (#29766)

kurokiasahi222 committed 1 year ago

Verified 0639034a

[docs] Remove redundant `-` and `the` from custom_tools.md (#29767)

windsonsea committed 1 year ago

Verified 5d1a58a6

[`BC 4.37 -> 4.38`] for Llama family, memory and speed (#29753)

ArthurZucker committed 1 year ago

Verified ff841900

[`BitsAndBytesConfig`] Warning for unused `kwargs` & safety checkers for `load_in_4bit` and `load_in_8bit` (#29761)

benjaminye committed 1 year ago

Verified 8dd4ce6f

Fix docker image build (#29762)

ydshieh committed 1 year ago

Verified 17e4467f

Update test reqs to include sentencepiece (#29756)

muellerzr committed 1 year ago

Verified c78f5772

Add LLaVa-1.6, bis (#29586)

NielsRogge committed 1 year ago

Verified d91fd7f9

Add correct batched handling for apply_chat_template (#29222)

Rocketknight1 committed 1 year ago

Verified 9d999481

SuperPointModel -> SuperPointForKeypointDetection (#29757)

amyeroberts committed 1 year ago

Verified 3c17c529

ArthurZucker committed 1 year ago

1248f092

Support sharded safetensors in TF (#29350)

Rocketknight1 committed 1 year ago

Verified 11ef35e8

fix jinja2 package version check (#29754)

Ricardo-L-C committed 1 year ago

Verified 870bbb4c

Update Mamba types and pass through use_cache attr to MambaModel (#29605)

koayon committed 1 year ago

Verified 76b3b20f

[Tests] Remove unused code (#29737)

NielsRogge committed 1 year ago

Verified 776c9d3a

fix galore layerwise with frozen params (#29743)

peterjc123 committed 1 year ago

Verified a1a74541

fixed the issue of DPO trainer that using one node and mutiple GPUs and set the device_map='auto' (#29695)

pengwei715 committed 1 year ago

Verified 8692aa88

Newer Older