huggingface/transformers

Pull Requests Commits

Llama 4 conversion fix for moe models

pcuenca committed 1 year ago

ca936ccf

Fix `qwen3_moe` tests (#38865)

ydshieh committed 1 year ago

Verified c77bcd88

🚨🚨 Fix initialization of Mask2Former (#38864)

Cyrilvallez committed 1 year ago

Verified 5a95ed5c

Fix `phi4_multimodal` tests (#38816)

ydshieh committed 1 year ago

Verified 309e8c96

enable misc test cases on XPU (#38852)

yao-matrix committed 1 year ago

Verified 3526e25d

Post-PR fixes! (#38868)

Rocketknight1 committed 1 year ago

Verified d058f81e

No more Tuple, List, Dict (#38797)

Rocketknight1 committed 1 year ago

Verified 508a7040

Update roc bert docs (#38835)

SohamPrabhu committed 1 year ago

Verified a396f432

Update CvT documentation with improved usage examples and additional … (#38731)

sezan92 committed 1 year ago

Verified 3ae52cc3

Add LightGlue model (#31718)

sbucaille committed 1 year ago

Verified e5a9ce48

Fix `qwen3` tests (#38862)

ydshieh committed 1 year ago

Verified 2507169b

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

simonreise committed 1 year ago

Verified 41e0c921

Fix `qwen2_5_vl` tests (#38845)

ydshieh committed 1 year ago

Verified c61ca64a

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

kimishpatel committed 1 year ago

Verified 37367c7d

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

Jingxiang-Zhang committed 1 year ago

Verified 9c878d2f

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

zucchini-nlp committed 1 year ago

Verified bf370e44

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

druvdub committed 1 year ago

Verified e61160c5

Updated aya_vision.md (#38749)

1himan committed 1 year ago

Verified 64e9b049

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

shawntan committed 1 year ago

Verified 5ab0f447

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

Vixel2006 committed 1 year ago

Verified a7593a1d

Fix broken tag in Longformer model card (#38828)

dross20 committed 1 year ago

Verified 18c7f32d

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

VolodymyrBg committed 1 year ago

Verified b44b04ee

Fix peft integration (#38841)

Cyrilvallez committed 1 year ago

Verified 93007286

add default mapping to peft integration

Cyrilvallez committed 1 year ago

Verified 60888496

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

ManuelFay committed 1 year ago

Verified ce6ac53a

Fix redundant code in Janus (#38826)

yaswanth19 committed 1 year ago

Verified 925da8ac

[internvl] fix video inference (#38811)

zucchini-nlp committed 1 year ago

Verified d2fd3868

Updated Albert model Card (#37753)

souvikchand committed 1 year ago

Verified d5d007a1

[docs] updated roberta model card (#38777)

allmight05 committed 1 year ago

Verified 443aafd3

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

stevhliu committed 1 year ago

Verified fdb5da59

Older