huggingface/transformers

Pull Requests Commits

ydshieh committed 1 year ago

e11d8a8b

ydshieh committed 1 year ago

265e0539

All models can be initialized on meta device (#37563)

Cyrilvallez committed 1 year ago

Verified 688f4707

Bridgetower fast image processor (#37373)

rootonchair committed 1 year ago

Verified 0a83588c

Fix Mamba2 Grouped SSD Support in the torch_forward Path (#37533)

cyang49 committed 1 year ago

Verified 40057300

Add EfficientNet Image PreProcessor (#37055)

zshn25 committed 1 year ago

Verified a7d2bbaa

[vlm] adjust max length for special tokens (#37342)

zucchini-nlp committed 1 year ago

Verified 32eca719

Fix pixel attention mask padding in smolvlm (#37497)

ManuelFay committed 1 year ago

Verified c94c59fc

Run `test_can_load_with_global_device_set` using a subprocess (#37553)

ydshieh committed 1 year ago

Verified 5a6de703

:red_circle: Update CLIP vision attention to new attention interface (#37498)

molbap committed 1 year ago

Verified 9a4ce647

Fix TimesFm doc issue (#37552)

Cyrilvallez committed 1 year ago

Verified dc822782

Make Ignored Columns ValueError More Informative (#33299)

wbuchanan committed 1 year ago

Verified 2f517200

Fix device issue for tapas (with `as_tensor`) (#37551)

ydshieh committed 1 year ago

Verified 0577cae8

docs(typo): Update ISSUES.md, fix a small typo (#37542)

NicholasWilsonDEV committed 1 year ago

Verified b33edf1b

add FlashAttentionKwargs and seq_idx to flat collator (#36456)

garrett361 committed 1 year ago

Verified 503541d7

Update quantization docs (#37439)

DerekLiu35 committed 1 year ago

Verified 9ddcf5fc

Add TimesFM Time Series Forecasting Model (#34082)

jinan-zhou committed 1 year ago

Verified a91020ae

Refactor torchao docs (#37490)

MekkCyber committed 1 year ago

Verified 8669c016

Keep Quark loading through meta device (#37538)

BowenBao committed 1 year ago

Verified e3d3b546

convert scale and zero to cuda when using HQQ backend (#37425)

phymhan committed 1 year ago

Verified 61436a93

Fixes hqq by following a new path for bias parameter in pre_quantized models (#37530)

MekkCyber committed 1 year ago

Verified 7752e748

More appropriate cuda warmup in resource-constrained hardware (#37550)

Cyrilvallez committed 1 year ago

Verified 7dafcd00

Add Fast Grounding-Dino Processor (#37108)

keetrap committed 1 year ago

Verified 6fd87d11

enable 6 rt_detr_v2 cases on xpu (#37548)

yao-matrix committed 1 year ago

Verified ed53809a

enable 3 mpt test cases on XPU (#37546)

yao-matrix committed 1 year ago

Verified d91858c2

Fix BitsAndBytesConfig JSON serialization in TrainingArguments (#37520)

astefanutti committed 1 year ago

Verified 4541c2cd

enable `test_offloaded_cache_implementation` on XPU (#37514)

yao-matrix committed 1 year ago

Verified a335dc4d

enable several cases on XPU (#37516)

yao-matrix committed 1 year ago

Verified 33f6c5a5

enable 5 cases on XPU (#37507)

yao-matrix committed 1 year ago

Verified 5ab7a7c6

Refactor ColPali model documentation (#37309)

Soum-Soum committed 1 year ago

Verified 3165eb7c

Older