huggingface/transformers

Pull Requests Commits

ArthurZucker committed 1 year ago

53fad641

Cache: revert DynamicCache init for BC (#33861)

gante committed 1 year ago

2fd49d2b

ArthurZucker committed 1 year ago

5df4ca82

fix post cherry-picking!

ArthurZucker committed 1 year ago

277ed58f

Fix Tensor + Embedding error in some cases when using SiglipVisionModel (#33994)

kaitolucifer committed 1 year ago

b1c237fc

properly fix and RUN_SLOW (#33965)

ArthurZucker committed 1 year ago

ae5f4916

Cache: revert DynamicCache init for BC (#33861)

gante committed 1 year ago

6ea04aaa

hot fix `self.position_embeddings->self.position_embedding` (#33958)

ArthurZucker committed 1 year ago

be968434

Ignore keys on `validate_rope` (#33753)

zucchini-nlp committed 1 year ago

333ec0a5

Config: lower `save_pretrained` exception to warning (#33906)

gante committed 1 year ago

3576fec8

Fixes for issue #33763 in idefics2 model (#33766)

aroun-coumar committed 1 year ago

f0686f56

remove warning v2 (#33761)

itazap committed 1 year ago

27f03e0a

manually fix PLBart tokenizer

ArthurZucker committed 1 year ago

e71a01a1

ArthurZucker committed 1 year ago

03178958

clean_up_tokenization_spaces=False if unset (#31938)

itazap committed 1 year ago

4ea1c43a

Generate: `can_generate()` recursive check (#33718)

gante committed 1 year ago

289edd9e

[`MllamaProcessor`] Update errors and API with multiple image (#33715)

ArthurZucker committed 1 year ago

c64be318

Release: v4.45.0

LysandreJik committed 1 year ago

2ef31dec

Add MLLama (#33703)

ArthurZucker committed 1 year ago

Verified 19d58d31

Add OmDet-Turbo (#31843)

yonigozlan committed 1 year ago

Verified 94f18cf2

Corrected max number for bf16 in transformer/docs (#33658)

alankashkash committed 1 year ago

Verified ade9e0fe

Add AdEMAMix optimizer (#33682)

matthewdouglas committed 1 year ago

Verified 196d35cc

Add SDPA support for M2M100 (#33309)

isaac091 committed 1 year ago

Verified 61e98cb9

Fix Megatron-LM tokenizer path (#33344)

hikari-kubota-fixstars committed 1 year ago

Verified 68049b17

HFQuantizer implementation for compressed-tensors library (#31704)

bfineran committed 1 year ago

Verified 574a9e12

fix code quality after merge

ArthurZucker committed 1 year ago

7e638ef2

[Pixtral] Improve docs, rename model (#33491)

NielsRogge committed 1 year ago

Verified 06e27e3d

bump tokenizers, fix added tokens fast (#32535)

ArthurZucker committed 1 year ago

Verified c6379858

tests: fix pytorch tensor placement errors (#33485)

dvrogozh committed 1 year ago

Verified 5e2916bc

🚨🚨 Setting default behavior of assisted decoding (#33657)

jmamou committed 1 year ago

Verified 52daf4ec

Older