huggingface/transformers

Pull Requests Commits

muellerzr committed 2 years ago

fd9b4a95

Even more TF test fixes (#28146)

Rocketknight1 committed 2 years ago

Verified 260b9d21

[`Mixtral` & `Mistral`] Add support for sdpa (#28133)

ArthurZucker committed 2 years ago

Verified f9a98c47

[Whisper] Use torch for stft if available (#26119)

sanchit-gandhi committed 2 years ago

Verified 814619f5

Fix `input_embeds` docstring in encoder-decoder architectures (#28168)

gante committed 2 years ago

Verified 7e93ce40

[bnb] Let's make serialization of 4bit models possible (#26037)

poedator committed 2 years ago

Verified 4f7806ef

disable test_retain_grad_hidden_states_attentions on SeamlessM4TModelWithTextInputTest (#28169)

dwyatte committed 2 years ago

Verified e268d7e5

Fix yolos resizing (#27663)

amyeroberts committed 2 years ago

Verified 1d777359

Generate: fix speculative decoding (#28166)

gante committed 2 years ago

Verified 45b70384

[docs] Trainer docs (#28145)

stevhliu committed 2 years ago

Verified 01c081d1

Align backbone stage selection with out_indices & out_features (#27606)

amyeroberts committed 2 years ago

Verified ee298a16

Update FA2 exception msg to point to hub discussions (#28161)

amyeroberts committed 2 years ago

Verified 224ab709

Avoid unnecessary warnings when loading `CLIPConfig` (#28108)

ydshieh committed 2 years ago

Verified 9924df9e

Fix weights not properly initialized due to shape mismatch (#28122)

ydshieh committed 2 years ago

Verified 7938c8c8

move code to Trainer.evaluate to enable use of that function with multiple datasets (#27844)

peter-sk committed 2 years ago

Verified 769a9542

[gpt-neox] Add attention_bias config to support model trained without attention biases (#28126)

dalgarak committed 2 years ago

Verified cd9f9d63

Fix FA2 integration (#28142)

pacman100 committed 2 years ago

Verified def581ef

Remove deprecated CPU dockerfiles (#28149)

ashahba committed 2 years ago

Verified b134f685

[docs] Fix mistral link in mixtral.md (#28143)

aaronjimv committed 2 years ago

Verified 38611086

Update modeling_utils.py (#28127)

mzelling committed 2 years ago

Verified 23f8e4db

[`Mixtral`] Fix loss + nits (#28115)

ArthurZucker committed 2 years ago

Verified 4a04b4cc

Generate: speculative decoding (#27979)

gante committed 2 years ago

Verified ac974199

Update split string in doctest to reflect #28087 (#28135)

amyeroberts committed 2 years ago

Verified bd7a3561

When save a model on TPU, make a copy to be moved to CPU (#27993)

qihqi committed 2 years ago

Verified 5aec50ec

[Doc] Fix token link in What 🤗 Transformers can do (#28123)

aaronjimv committed 2 years ago

Verified 4edffda6

Fix a typo in tokenizer documentation (#28118)

mssalvatore committed 2 years ago

Verified c52b515e

[docs] General doc fixes (#28087)

stevhliu committed 2 years ago

Verified a52e180a

Fix indentation error - semantic_segmentation.md (#28117)

rajveer43 committed 2 years ago

Verified 08a6e7a7

More TF fixes (#28081)

Rocketknight1 committed 2 years ago

Verified 71d47f0a

Remove warning if `DISABLE_TELEMETRY` is used (#28113)

Wauplin committed 2 years ago

Verified 0695b242

Older