huggingface/transformers

Pull Requests Commits

[pagged-attention] fix off-by-1 error in pagged attention generation (#39258)

kashif committed 1 year ago

Verified db05e4ff

[CI] fix docs (#39273)

gante committed 1 year ago

Verified 6f1a4389

Add Aimv2 model (#36625)

yaswanth19 committed 1 year ago

Verified fbdaa7b0

Add Doge model (#35891)

LoserCheems committed 1 year ago

Verified d8590b4b

Fix errors when use verl to train GLM4.1v model (#39199)

kaln27 committed 1 year ago

Verified d370bc64

fix recompiles due to instance key, and deepcopy issues (#39270)

ArthurZucker committed 1 year ago

Verified 5fb8bb3e

fix(generation): stop beam search per-instance when heuristic satisfied (#38778)

guang-yng committed 1 year ago

Verified 356fd681

remove broken block (#39255)

molbap committed 1 year ago

Verified 0b0ede8b

Skip `test_eager_matches sdpa generate` and update an integration test for blip-like models (#39248)

ydshieh committed 1 year ago

Verified a21557fa

Fix license text, duplicate assignment, and typo in constant names (#39250)

gudwls215 committed 1 year ago

Verified ea3c2c02

fix xpu failures on PT 2.7 and 2.8 w/o IPEX and enable hqq cases on XPU (#39187)

yao-matrix committed 1 year ago

Verified b2816da8

Glm 4 doc (#39247)

zRzRzRzRzRzRzR committed 1 year ago

Verified 17b3c96c

Update LED model card (#39233)

dross20 committed 1 year ago

Verified bbca9782

fix some flaky tests in `tests/generation/test_utils.py` (#39254)

ydshieh committed 1 year ago

Verified 41e865bb

Simplify Mixtral and its modular children (#39252)

Cyrilvallez committed 1 year ago

Verified 93747d89

Add `segmentation_maps` support to MobileNetV2ImageProcessor (#37312)

simonreise committed 1 year ago

Verified 3993ee1e

Clarify per_device_train_batch_size scaling in TrainingArguments (#38… (#38857)

Shohail-Ismail committed 1 year ago

Verified b96f213f

Add Korean translation for glossary.md (#38804)

JoosunH committed 1 year ago

Verified 96980525

Update tiny-agents example (#39245)

Wauplin committed 1 year ago

Verified bf203aa9

adjust input and output texts for test_modeling_recurrent_gemma.py (#39190)

kaixuanliu committed 1 year ago

Verified c4e39ee5

enable xpu on kv-cache and hqq doc (#39246)

jiqing-feng committed 1 year ago

Verified 14cba7ad

Fix patch helper (#39216)

Cyrilvallez committed 1 year ago

Verified 32db48db

RotaryEmbeddings change `is not None` -> `isinstance(..., dict)` (#39145)

qubvel committed 1 year ago

Verified a3618d48

fix `fastspeech2_conformer` tests (#39229)

ydshieh committed 1 year ago

Verified 9b09fe47

[bugfix] fix flash attention 2 unavailable error on Ascend NPU (#39166)

FightingZhen committed 1 year ago

Verified 00e9efce

[modular] Simplify logic and docstring handling (#39185)

Cyrilvallez committed 1 year ago

Verified 056fa73f

Make _compute_dynamic_ntk_parameters exportable (#39171)

xadupre committed 1 year ago

Verified f16fbfb8

fix bug using FSDP V1 will lead to model device not properly set (#39177)

kaixuanliu committed 1 year ago

Verified 4243bb84

Don't send new comment if the previous one is less than 30 minutes (unless the content is changed) (#39170)

ydshieh committed 1 year ago

Verified 34c16167

fix typo in Gemma3n notes (#39196)

davanstrien committed 1 year ago

Verified b8f397e4

Newer Older