NielsRogge/transformers

Pull Requests Commits

allow using tokenizer.pad as a collate_fn in pytorch

thomwolf committed 5 years ago

22bd22fc

bart-large-mnli-yahoo-answers model card (#6133)

joeddav committed 5 years ago

Verified 8edfaaa8

Model output test (#6155)

sgugger committed 5 years ago

Verified d951c14a

Harmonize both Trainers API (#6157)

sgugger committed 5 years ago

Verified 86caab1e

readme m3hrdadfi/albert-fa-base-v2 (#6153)

m3hrdadfi committed 5 years ago

Verified 603cd81a

parse arguments from dict (#4869)

patil-suraj committed 5 years ago

Verified 838dc06f

Replace mecab-python3 with fugashi for Japanese tokenization (#6086)

polm committed 5 years ago

Verified cf3cf304

enable easy checkout switch (#5645)

stas00 committed 5 years ago

Verified f250beb8

Create README.md (#6169)

kolk committed 5 years ago

Verified 7d50af4b

Add Pytorch Native AMP support in Trainer (#6151)

prajjwal1 committed 5 years ago

Verified 0034a1d2

Enable ONNX/ONNXRuntime optimizations through converter script (#6131)

mfuntowicz committed 5 years ago

Verified 7231f7b5

correct the correction (#6163)

stas00 committed 5 years ago

Verified c0b93a1c

stas00 committed 5 years ago

Verified a2f6d521

Doc tokenizer (#6110)

sgugger committed 5 years ago

Verified f3065abd

Addition of a DialoguePipeline (#5516)

guillaume-be committed 5 years ago

Verified e642c789

Fix FlauBERT GPU test (#6142)

LysandreJik committed 5 years ago

Verified ec026747

Switch from return_tuple to return_dict (#6138)

sgugger committed 5 years ago

Verified 91cb9546

Tf trainer cleanup (#6143)

sgugger committed 5 years ago

Verified 562b6369

add another e.g. to avoid confusion (#6055)

orena1 committed 5 years ago

Verified c127d055

Actually the extra_id are from 0-99 and not from 1-100 (#5967)

orena1 committed 5 years ago

Verified d24ea708

[s2s] add support for overriding config params (#6149)

stas00 committed 5 years ago

Verified 3212b885

Rework TF trainer (#6038)

jplu committed 5 years ago

Verified 54f9fbef

[WIP] Test TF Flaubert + Add {XLM, Flaubert}{TokenClassification, MultipleC… (#5614)

LysandreJik committed 5 years ago

Verified 3f94170a

Use google style to document properties (#6130)

sgugger committed 5 years ago

Verified 8a8ae276

Fix TF CTRL model naming (#6134)

jplu committed 5 years ago

Verified fc64559c

XLNet PLM Readme (#6121)

LysandreJik committed 5 years ago

Verified 641b873c

add deepset/xlm-roberta-large-squad2 model card (#6128)

Timoeller committed 5 years ago

Verified 8d157c93

Added capability to quantize a model while exporting through ONNX. (#6089)

mfuntowicz committed 5 years ago

Verified 6c002853

Use FutureWarning to deprecate (#6111)

sgugger committed 5 years ago

Verified 25de74cc

ONNX documentation (#5992)

mfuntowicz committed 5 years ago

Verified 640550fc

Older