NielsRogge/transformers

Pull Requests Commits

[docs] Custom semantic segmentation dataset (#27859)

stevhliu committed 2 years ago

Verified f7595760

Generate: All logits processors are documented and have examples (#27796)

gante committed 2 years ago

Verified 58e7f9bb

Fix TF loading PT safetensors when weights are tied (#27490)

Rocketknight1 committed 2 years ago

Verified 47500b1d

Show new failing tests in a more clear way in slack report (#27881)

ydshieh committed 2 years ago

Verified 9f1f11a2

Fix device of masks in tests (#27887)

fxmarty committed 2 years ago

Verified c99f2547

update version of warning notification for `get_default_device` to v4.38 (#27848)

ji-huazhong committed 2 years ago

Verified fc71e815

update `create_model_card` to properly save peft details when using Trainer with PEFT (#27754)

pacman100 committed 2 years ago

Verified 5324bf9c

Allow `# Ignore copy` (#27328)

ydshieh committed 2 years ago

Verified 52746922

[`Llava`] Add Llava to transformers (#27662)

younesbelkada committed 2 years ago

Verified 44b5506d

fix: fix gradient accumulate step for learning rate (#27667)

pphuc25 committed 2 years ago

Verified 0410a29a

[`FA-2`] Add Flash Attention to `Phi` (#27661)

susnato committed 2 years ago

Verified f84d85ba

[i18n-fr] Translate autoclass tutorial to French (#27659)

NoB0 committed 2 years ago

Verified 06f56168

Fix bug of _prepare_4d_attention_mask (#27847)

jiqing-feng committed 2 years ago

Verified 4d806dba

Add Llama Flax Implementation (#24587)

vvvm23 committed 2 years ago

Verified 75336c17

Fix beam score calculation issue for JAX version (#27816)

VsonicV committed 2 years ago

Verified 7fc80724

Translating en/model_doc folder docs to Japanese(from `blip` to `clap`) 🇯🇵 (#27673)

rajveer43 committed 2 years ago

Verified 9660e27c

[`Flash Attention 2`] Add flash attention 2 for GPT-Neo-X (#26463)

younesbelkada committed 2 years ago

Verified 9270ab08

Avoid class attribute `_keep_in_fp32_modules` being modified (#27867)

ydshieh committed 2 years ago

Verified 87714b3d

removed the delete doc workflows (#27852)

MKhalusova committed 2 years ago

Verified d6392482

Update CUDA versions for DeepSpeed (#27853)

muellerzr committed 2 years ago

Verified acd65316

[`Docs`] Update broken image on fused modules (#27856)

younesbelkada committed 2 years ago

Verified ba52dec4

Documentation: Spanish translation of perplexity.mdx (#27807)

aaronjimv committed 2 years ago

Verified da1d0d40

fix(whisper): mutable generation config (#27833)

badayvedat committed 2 years ago

Verified 788730c6

Update `VitDetModelTester.get_config` to use `pretrain_image_size` (#27831)

ydshieh committed 2 years ago

Verified ac975074

⚠️ [VitDet] Fix test (#27832)

NielsRogge committed 2 years ago

Verified 28e2887a

[Time series] Add PatchTSMixer (#26247)

ajati committed 2 years ago

Verified b242d0f2

Move tensors to same device to enable IDEFICS naive MP training (#27746)

willemsenbram committed 2 years ago

Verified e5c12c03

[`ClipVision`] `accelerate` support for clip-vision (#27851)

younesbelkada committed 2 years ago

Verified 3e68944c

Generate: Update VisionEncoderDecoder test value (#27850)

gante committed 2 years ago

Verified b7e6d120

Faster generation using AWQ + Fused modules (#27411)

younesbelkada committed 2 years ago

Verified fdb85be4

Newer Older