huggingface/transformers

Pull Requests Commits

ArthurZucker committed 27 days ago

4018a25b

add XPU Expectations for `higgs_audio_v2` tests (#44482)

kaixuanliu committed 27 days ago

Verified 0e3be387

chameleon added to MODELS_WITH_INCORRECT_HUB_TOKENIZER_CLASS (#44475)

itazap committed 27 days ago

Verified 52c93cfb

[CB] Add paged_attention kernel (#44379)

remi-or committed 27 days ago

Verified a08aa52d

Revert "test merge queue 1" (#44552)

ydshieh committed 28 days ago

Verified 2da98fda

test merge queue 1 (#44529)

ydshieh2 committed 28 days ago

Verified f428423e

Fix: Remove references to `text2text-generation`, `summarization` and `translation` pipeline tasks (#44510)

math-hiyoko committed 28 days ago

Verified 138c1c81

fix(testing): Fix MoonshineEncoder UnboundLocalError and Florence2VisionBackbone dtype mismatch (#44503)

harshaljanjani committed 28 days ago

Verified 1a50a3b1

fix FSDP loading with meta devices (#44473)

winglian committed 28 days ago

Verified 0a0ac7a2

Fix: Remove references to transformers run command (#44513)

math-hiyoko committed 28 days ago

Verified a049c004

[LW-DETR] Fix training (#44441)

NielsRogge committed 28 days ago

Verified a50b099a

fix(tokenizer): Only strip Fast from class names in AutoTokenizer if used as a suffix (#44443)

harshaljanjani committed 28 days ago

Verified f8786de5

[tie weights] 🚨 If both weights are present with same weights, still tie them (#44497)

Cyrilvallez committed 28 days ago

Verified 09677910

Fix AMD Docker image build timeout by pinning Flash Attention commit (#44546)

Abdennacer-Badaoui committed 28 days ago

Verified ce7efd8f

Make `_prepare_input_fn` and `_prepare_output_fn` instance methods (#44499)

michaelbenayoun committed 28 days ago

Verified 86eafab5

Fix: Conditionally import `torch.distributed.fsdp` in `trainer_seq2seq.py` (#44507)

0xDELUXA committed 28 days ago

Verified adc0d9ae

Fix ShieldGemma2 non-reproducible outputs by adding _tied_weights_keys (#44358)

hardikmeisheri committed 28 days ago

Verified 7d6fa935

Tensor Parallelism and `mps` device (#44506)

michaelbenayoun committed 28 days ago

Verified 78aee7e8

Fix failing `GPTNeoModelLanguageGenerationTest` (#44515)

Sai-Suraj-27 committed 28 days ago

Verified 36279211

Fix failing `MarianIntegrationTests` (#44519)

Sai-Suraj-27 committed 28 days ago

Verified 395a5ed3

fix pin_memory for contiguous batching (#44455)

jiqing-feng committed 28 days ago

Verified 1cdae08e

Fix continuous batching for multimodal models (#44436)

jw9603 committed 28 days ago

Verified 70ca366f

Fix KeyError in _parse_type_hint when Union contains Any (#44525)

jnMetaCode committed 28 days ago

Verified 0130c7da

Fix AssistantTracker.is_active() returning False after activation with empty lists (#44524)

jnMetaCode committed 28 days ago

Verified deb7dcd9

Fix and re-enable extra_state tests (#43510)

pstjohn committed 28 days ago

Verified b8d6b845

Fix ansi codes in loading reports when not connected to terminal (#44544)

Cyrilvallez committed 28 days ago

Verified 24ba4c84

Follow-up typing checking fixes (#44500)

tarekziade committed 28 days ago

Verified 976d8cc9

Supplement skip logic for XPU in the CPU-only tp tests (#44536)

YangKai0616 committed 28 days ago

Verified de0c82d2

Fix backend dependency (#44542)

Cyrilvallez committed 28 days ago

Verified 6a9b2c0c

Add a new job in `build_pr_documentation.yml` (will be the new required job) (#44538)

ydshieh committed 28 days ago

Verified 70162852

Older