huggingface/transformers

Pull Requests Commits

Fix typos in docstrings, comments, and error messages (#43949)

DimiChatzipavlis committed 198 days ago

Verified c8f112d4

Fix `get_number_of_image_tokens` (#43948)

zucchini-nlp committed 198 days ago

Verified 53a49c48

Fix qwen-vl position ids when generating several times (#43952)

zucchini-nlp committed 198 days ago

Verified 3b8f0948

Timm backbone saves and loads `out_features` (#43886)

zucchini-nlp committed 198 days ago

Verified 0aebdc5c

fix: Better weight decay exclusion in `run_*_no‑trainer.py` examples (#43947)

casinca committed 198 days ago

Verified 2a232cd9

Fix weight decay exclusions in `run_*_no‑trainer.py` examples (#42769)

casinca committed 198 days ago

Verified 2caa05de

Modernize string formatting (f-strings) in conversion scripts (#43943)

DimiChatzipavlis committed 198 days ago

Verified a4fdaa5f

Add explicit utf-8 encoding to CircleCI scripts for Windows compatibility (#43925)

DimiChatzipavlis committed 198 days ago

Verified f013b342

Fix flaky test for multimodal LLMs (#43944)

Rocketknight1 committed 198 days ago

Verified 32def650

Fix docker files (#43946)

ydshieh committed 198 days ago

Verified 85147681

Fix bark test (#43942)

Cyrilvallez committed 198 days ago

Verified 0afacc2a

Fix lifted_tensor in Gemma3n export which dynamo can't reason about (#43801)

robell committed 198 days ago

Verified d5131eb2

Prepare and keep track of position ids in `generate` (#43734)

zucchini-nlp committed 198 days ago

Verified 9d9b012d

add openAI style `image_url` content support in `apply_chat_template` (#43786)

kaixuanliu committed 198 days ago

Verified 3a87a7a5

Fix AttributeError in OwlViT conversion script for Python 3.10+ (#43922)

DimiChatzipavlis committed 198 days ago

Verified 4d5d49c3

Revert #43897 (#43923)

Rocketknight1 committed 199 days ago

Verified ae05b2ae

fix: TextClassificationPipeline docs mentioning deprecated return_all_scores (#43903)

math-hiyoko committed 199 days ago

Verified e6fa62bf

Harmonize `input_embeds` to `inputs_embeds` everywhere (#43916)

Cyrilvallez committed 199 days ago

Verified 6c6f70cb

Ernie4 5 vl moe (#43755)

kaixuanliu committed 199 days ago

Verified ac6cba66

Changes to cache_utils should trigger all tests all the time (#43920)

Cyrilvallez committed 199 days ago

Verified 48295ecb

Move `_keys_to_ignore_on_load_missing` for now (#43893)

ArthurZucker committed 199 days ago

Verified 64e41924

Fix old tech stack in doc (#43902)

cyyever committed 199 days ago

Verified b52b6631

Fix missing fast image patch counter in Glm46V (#43877)

weiguangli-io committed 199 days ago

Verified 7d4c298b

fix(moe): Handle dtype mismatch in torch._grouped_mm with autocast (#43839)

Mr-Neutr0n committed 199 days ago

Verified 007bb8c3

Train sampler unification (#43138)

jiosephlee committed 199 days ago

Verified 372c27e7

revert place_model_on_device to property (#43895)

SunMarc committed 199 days ago

Verified 8efd8151

Decorate cache updates with no_grad, just in case (#43897)

Rocketknight1 committed 199 days ago

Verified 6c4710bd

Update KERNELS_MIN_VERSION to 0.10.2 to be the same as setup.py (#43753)

cyyever committed 199 days ago

Verified a316807d

Fix typo in quantization_operations in PEFT integrations (#43821)

redpanda1995 committed 199 days ago

Verified 42dc276c

adding BC for custom toks accessing slow tok attrs deprecated in v5 (#43898)

itazap committed 200 days ago

Verified e2372ffa

Newer Older