huggingface/transformers

Pull Requests Commits

muellerzr committed 327 days ago

f8a963c1

muellerzr committed 327 days ago

f839aa20

muellerzr committed 327 days ago

919bcbec

Fix OS err (#36094)

muellerzr committed 328 days ago

Verified ba29a439

Move audio top_k tests to the right file and add slow decorator (#36072)

Rocketknight1 committed 328 days ago

Verified a18b7fdd

Fix bug in apply_rotary_pos_emb_flashatt: in Qwen2-5-VL (#36065)

DeepWaved committed 328 days ago

Verified 014047e1

Adding RT-DETRv2 for object detection (#34773)

jadechoghari committed 328 days ago

Verified 006d9249

[docs] fix outdated example code in `trainer.md` (#36066)

faaany committed 328 days ago

Verified 6246c032

Fix StopStringCriteria to handle tokens above len(tokenizer) (#35797)

Rocketknight1 committed 328 days ago

Verified 4563ba2c

Fix model kwargs (#35875)

muellerzr committed 329 days ago

Verified 28f73bc3

Fix words typos in ggml test. (#36060)

zhanluxianshen committed 329 days ago

Verified 1590c664

Nail in edge case of torch dtype being overriden permantly in the case of an error (#35845)

muellerzr committed 329 days ago

Verified 1ce0e299

Save checkpoint to temporary directory to handle partial saves during failures (#35580)

SilverSoldier committed 329 days ago

Verified e3458af7

Paligemma: fix generation with Gemma2 (#36044)

zucchini-nlp committed 329 days ago

Verified 3dd1de39

Update `test_flash_attn_2_can_dispatch_composite_models` (#36050)

ydshieh committed 329 days ago

Verified dce99708

Fix repo consistency (#36063)

ydshieh committed 329 days ago

Verified 37faa97d

Fix usage of unpad_input function (#35925)

pavelgein committed 329 days ago

Verified ed98ad35

Iterative generation using Input embeds and `past_key_values` (#35890)

yaswanth19 committed 329 days ago

Verified 7aee036e

Add `Qwen2VLImageProcessorFast` into `Qwen2VLProcessor` (#35987)

yeliudev committed 329 days ago

Verified b5f327f3

Fix Audio Classification Pipeline top_k Documentation Mismatch and Bug #35736 (#35771)

sambhavnoobcoder committed 330 days ago

Verified 0de15c98

Fix how we compute the final non-padding token for ForSequenceClassification models (#35911)

Rocketknight1 committed 330 days ago

Verified 694aaa7f

[docs] no hard-coding cuda (#36043)

faaany committed 330 days ago

Verified 531d1511

[docs] fix bugs in the bitsandbytes documentation (#35868)

faaany committed 330 days ago

Verified 7399f802

[docs] no hard coding cuda as bnb has multi-backend support (#35867)

faaany committed 330 days ago

Verified 0a1a8e3c

DeepSpeed github repo move sync (#36021)

stas00 committed 330 days ago

Verified 9dc1efa5

add support for empty list as input to create_model_card (#36042)

ROZBEH committed 330 days ago

Verified c772bff3

Add XPU type for work-around -inf mask causing sdpa NaN issue in modeling files (#35647)

Liangliang-Ma committed 330 days ago

Verified 315a9f49

Fix synced multi-GPU generation with LLMs and VLMs (#35893)

ManukyanD committed 330 days ago

Verified d8080d55

Fix Gemma2 synced multi-GPU generation (#35232)

ManukyanD committed 330 days ago

Verified 4831a94e

Refactoring of ImageProcessorFast (#35069)

yonigozlan committed 330 days ago

Verified fa56dcc2

Older