huggingface/transformers

Pull Requests Commits

MekkCyber committed 1 day ago

ee40de9c

Merge branch 'main' into add-mlx-quantization

MekkCyber committed 1 day ago

Verified 58a58ed3

MekkCyber committed 1 day ago

b247031c

MekkCyber committed 1 day ago

4fb1ddfb

MekkCyber committed 1 day ago

adb51ade

Revert #43897 (#43923)

Rocketknight1 committed 1 day ago

Verified ae05b2ae

fix: TextClassificationPipeline docs mentioning deprecated return_all_scores (#43903)

math-hiyoko committed 1 day ago

Verified e6fa62bf

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

Cyrilvallez committed 1 day ago

Verified 6c6f70cb

Ernie4 5 vl moe (#43755)

kaixuanliu committed 1 day ago

Verified ac6cba66

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

Cyrilvallez committed 1 day ago

Verified 48295ecb

Move `_keys_to_ignore_on_load_missing` for now (#43893)

ArthurZucker committed 1 day ago

Verified 64e41924

Fix old tech stack in doc (#43902)

cyyever committed 1 day ago

Verified b52b6631

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

OiPunk committed 1 day ago

Verified 7d4c298b

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

Mr-Neutr0n committed 1 day ago

Verified 007bb8c3

Train sampler unification (#43138)

jiosephlee committed 1 day ago

Verified 372c27e7

revert place_model_on_device to property (#43895)

SunMarc committed 1 day ago

Verified 8efd8151

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

Rocketknight1 committed 2 days ago

Verified 6c4710bd

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

cyyever committed 2 days ago

Verified a316807d

Fix typo in quantization_operations in PEFT integrations (#43821)

redpanda1995 committed 2 days ago

Verified 42dc276c

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

itazap committed 2 days ago

Verified e2372ffa

Minor changes trainer (#43744)

SunMarc committed 2 days ago

Verified 24c743e3

Fix expected DAC outputs due to (old) change in CI settings. (#43896)

ebezzam committed 2 days ago

Verified cc88192b

Remove mask slicing in all eager attentions (#42186)

Cyrilvallez committed 2 days ago

Verified e00bac44

Separate `check_model_inputs` into `capture_outputs` and `merge_with_config_defaults` + ensure correctness (#43862)

Cyrilvallez committed 2 days ago

Verified 42791a34

[fix] DAC model: Apply STE in Dac.from_latents to match the forward pass (#43820)

harshaljanjani committed 2 days ago

Verified 3d24ffaa

Update hub metadata (#43892)

zucchini-nlp committed 2 days ago

Verified 15bada75

[Docs] Add docs for GLM-OCR and fix EomT-DINOv3 (#43710)

NielsRogge committed 2 days ago

Verified 60a5dbaa

[`fix`] Use `last_hidden_state` key from `get_image_features` for llama4 (#43882)

tomaarsen committed 2 days ago

Verified 71d7fc58

Refactor trainer init (#43807)

SunMarc committed 2 days ago

Verified e12aa2d2

fix fbgemm fp8 multi-device load failure. (#43581)

sywangyi committed 2 days ago

Verified 6d86e9a3

Older