huggingface/transformers

Pull Requests Commits

Allow generic composite models to pass more kwargs (#24927)

ydshieh committed 2 years ago

Verified 1e662f0f

🌐 [i18n-KO] Translated `perf_infer_cpu.md` to Korean (#24920)

junejae committed 2 years ago

Verified b51312e2

[DOCS] add example NoBadWordsLogitsProcessor (#25046)

SoyGema committed 2 years ago

Verified b99f7bd4

[`MPT`] Add MosaicML's `MPT` model to transformers (#24629)

ArthurZucker committed 2 years ago

Verified dcb183f4

Fix: repeat per sample for SAM image embeddings (#25074)

xk-huang committed 2 years ago

Verified 1dbc1440

🌐 [i18n-KO] Translated `hpo_train.md` to Korean (#24968)

harheem committed 2 years ago

Verified cb8abee5

[`generate`] Only warn users if the `generation_config`'s `max_length` is set to the default value (#25030)

ArthurZucker committed 2 years ago

Verified f2c1df93

replace `per_gpu_eval_batch_size` with `per_device_eval_batch_size` in readme of multiple-choice task (#25078)

ji-huazhong committed 2 years ago

Verified c879318c

Fix broken link in README_hd.md (#25067)

susnato committed 2 years ago

Verified 25e443c0

Set `TF32` flag for PyTorch cuDNN backend (#25075)

XuehaiPan committed 2 years ago

Verified 6bc61aa7

fix: add TOC anchor link (#25066)

eenzeenee committed 2 years ago

Verified 5dba88b2

Fix last models for common tests that are too big. (#25058)

sgugger committed 2 years ago

Verified f295fc8a

🌐 [i18n-KO] Translated `perf_hardware.md` to Korean (#24966)

augustinLib committed 2 years ago

Verified ee1eb3b3

🌐 [i18n-KO] Translated `<tf_xla>.md` to Korean (#24904)

54data committed 2 years ago

Verified f6fe1d55

[Docs] fix rope_scaling doc string (#25072)

kashif committed 2 years ago

Verified faf25c04

Generate - add beam indices output in contrained beam search (#25042)

gante committed 2 years ago

Verified c0742b15

[`RWKV`] Add note in doc on `RwkvStoppingCriteria` (#25055)

ArthurZucker committed 2 years ago

Verified c53a6eae

Better error message when signal is not supported on OS (#25049)

sgugger committed 2 years ago

Verified d2295708

🌐 [i18n-KO] Translated `perf_train_cpu.md` to Korean (#24911)

seank021 committed 2 years ago

Verified c0d1c330

[`8bit`] Fix 8bit corner case with Blip2 8bit (#25047)

younesbelkada committed 2 years ago

Verified b08f41e6

compute_loss in trainer failing to label shift for PEFT model when label smoothing enabled. (#25044)

njbrake committed 2 years ago

Verified 3611fc90

Pvt model (#24720)

Xrenya committed 2 years ago

Verified a03d13c8

Comment again print statement

sgugger committed 2 years ago

Verified afe8bfc0

Make more test models smaller (#25005)

sgugger committed 2 years ago

Verified 42571f6e

Fix typo in LlamaTokenizerFast docstring example (#25018)

sbrunk committed 2 years ago

Verified 8f1f0bf5

Add dispatch_batches to training arguments (#25038)

muellerzr committed 2 years ago

Verified 3b734f50

🌐 [i18n-KO] Translated `testing.md` to Korean (#24900)

Sunmin0520 committed 2 years ago

Verified 9d2b983e

🌐[i18n-KO] Translated performance.md to Korean (#24883)

augustinLib committed 2 years ago

Verified 383be1b7

Better handling missing SYS in llama conversation tokenizer (#24997)

ichernev committed 2 years ago

Verified efb2ba66

Support GatedRepoError + use raise from (#25034)

Wauplin committed 2 years ago

Verified 67049231

Newer Older