NielsRogge/transformers

Pull Requests Commits

NielsRogge committed 1 year ago

910d11bd

NielsRogge committed 1 year ago

5f56c143

Merge remote-tracking branch 'upstream/main' into feature/update_idefics_ignore_index

NielsRogge committed 1 year ago

758345a2

`tokenizer_class = "AutoTokenizer"` Llava Family (#30912)

ArthurZucker committed 1 year ago

Verified bb48e921

Fix a shape annotation and typos in `mamba` slow forward (#30691)

vasqu committed 1 year ago

Verified 76e05301

Add AutoFeatureExtractor support to Wav2Vec2ProcessorWithLM (#28706)

ylacombe committed 1 year ago

Verified e6708709

fix for custom pipeline configuration (#29004)

not-lain committed 1 year ago

Verified c11ac785

separate kwargs in processor (similar to #30193) (#30905)

Eric2i committed 1 year ago

Verified 7b4b4564

Fix num_hidden_layers in initialization of new model in Mamba (#30403)

SrGonao committed 1 year ago

Verified 18349164

add return_token_timestamps to WhisperProcessor (#30812)

kamilakesbi committed 1 year ago

Verified 1c2bb3ac

DeformableDETR two stage support bfloat16 (#30907)

DonggeunYu committed 1 year ago

Verified 66b0d9ee

LLaVa-Next: Update docs with batched inference (#30857)

zucchini-nlp committed 1 year ago

Verified 5d0bf59b

Add support for torch.compile dynamic shapes (#30560)

warner-benjamin committed 1 year ago

Verified cd6bd0af

FIX / Quantization: Fix Dockerfile build (#30890)

younesbelkada committed 1 year ago

Verified fce78fd0

Add TokenClassification for Mistral, Mixtral and Qwen2 (#29878)

josephenguehard committed 1 year ago

Verified 07bf2dff

Update ignore index

NielsRogge committed 1 year ago

4b850078

Enable dynamic resolution input for Swin Transformer and variants (#30656)

the-neural-networker committed 1 year ago

Verified 481a9578

ArthurZucker committed 1 year ago

b6eb708b

Add fixed resize and pad strategy for object detection (#30742)

qubvel committed 1 year ago

Verified bf646fbf

update release script (#30880)

ArthurZucker committed 1 year ago

Verified e9a8041d

Support arbitrary processor (#30875)

ArthurZucker committed 1 year ago

Verified 0a9300f4

[whisper] fix multilingual fine-tuning (#30865)

sanchit-gandhi committed 1 year ago

Verified 57edd84b

Fix dependencies for image classification example (#30842)

jla524 committed 1 year ago

Verified 977ce58a

Enable device map (#30870)

darshana1406 committed 1 year ago

Verified 3802e786

Remove deprecated logic and warnings (#30743)

amyeroberts committed 1 year ago

Verified 57c965a8

TEST: Add llama logits tests (#30835)

younesbelkada committed 1 year ago

Verified 3d7d3a87

Fix VideoLlava imports (#30867)

amyeroberts committed 1 year ago

Verified 15c74a28

TST / Quantization: Reverting to torch==2.2.1 (#30866)

younesbelkada committed 1 year ago

Verified 4e17e7dc

Docs: update example with assisted generation + sample (#30853)

gante committed 1 year ago

Verified f4014e75

Video-LLaVa: Fix docs (#30855)

zucchini-nlp committed 1 year ago

Verified 95b3c381

Older