huggingface/transformers

Pull Requests Commits

McPatate committed 303 days ago

Verified fb43e6d8

feat: add sliding window attention support

McPatate committed 303 days ago

Verified f0b2ff7e

Fix `qwen3` tests (#38862)

ydshieh committed 303 days ago

Verified 2507169b

Improve `auxiliary_in_channels` default behavior in UperNet (#37540)

simonreise committed 303 days ago

Verified 41e0c921

Fix `qwen2_5_vl` tests (#38845)

ydshieh committed 304 days ago

Verified c61ca64a

Allow customization of sdpa in executorch.py (#38827)

kimishpatel committed 304 days ago

Verified 37367c7d

Fix incorrect width ratio calculation in Llama4 image processor (#38842)

Jingxiang-Zhang committed 304 days ago

Verified 9c878d2f

[video processor] fix BC when no video config if found (#38840)

zucchini-nlp committed 304 days ago

Verified bf370e44

Remove merge conflict artifacts in Albert model doc (#38849)

druvdub committed 304 days ago

Verified e61160c5

Updated aya_vision.md (#38749)

1himan committed 304 days ago

Verified 64e9b049

GraniteMoeHybrid: Allow for only shared expert case. (#38801)

shawntan committed 304 days ago

Verified 5ab0f447

[BugFix] QA pipeline edge case: `align_to_words=True` in `QuestionAnsweringPipeline` can lead to duplicate answers (#38761)

Vixel2006 committed 304 days ago

Verified a7593a1d

Fix broken tag in Longformer model card (#38828)

dross20 committed 304 days ago

Verified 18c7f32d

Fix broken notebooks link in Italian training docs (#38834)

VolodymyrBg committed 304 days ago

Verified b44b04ee

Fix peft integration (#38841)

Cyrilvallez committed 305 days ago

Verified 93007286

add default mapping to peft integration

Cyrilvallez committed 305 days ago

Verified 60888496

bugfix: propage weight key_mapping to peft to fix 3.52 VLM renaming (#38627)

ManuelFay committed 305 days ago

Verified ce6ac53a

Fix redundant code in Janus (#38826)

yaswanth19 committed 305 days ago

Verified 925da8ac

[internvl] fix video inference (#38811)

zucchini-nlp committed 305 days ago

Verified d2fd3868

Updated Albert model Card (#37753)

souvikchand committed 307 days ago

Verified d5d007a1

[docs] updated roberta model card (#38777)

allmight05 committed 307 days ago

Verified 443aafd3

[docs] Update docs moved to the course (#38800)

stevhliu committed 307 days ago

Verified fdb5da59

fixed docstring in modular_qwen2_5_vl.py (#38798)

lawrencefeng17 committed 307 days ago

Verified 8b737995

Add V-JEPA for video classification model (#38788)

qubvel committed 307 days ago

Verified 9bec2654

Fix trainer.py not showing signature columns (#38465)

nenesekai committed 307 days ago

Verified 2ff964bc

Fix a minor security issue (#38815)

ydshieh committed 307 days ago

Verified 4c3c177e

change fsdp_strategy to fsdp in TrainingArguments in accelerate doc (#38807)

PT-10 committed 307 days ago

Verified 93445aed

Refactor DBRX tests to use CausalLMModelTest base classes (#38475)

Rocketknight1 committed 307 days ago

Verified b82a45b3

Use `wandb.run.url` instead of `wandb.run.get_url()` (deprecated) (#38817)

qgallouedec committed 307 days ago

Verified 64041694

Expectation fixes and added AMD expectations (#38729)

remi-or committed 307 days ago

Verified 9ff246db

Older