NielsRogge/transformers

Pull Requests Commits

add model_save_dir argument

thomwolf committed 6 years ago

9b87c341

load from pretrained model if no saving during training

thomwolf committed 6 years ago

cbf09af9

only main script save in distributed ops

thomwolf committed 6 years ago

9d56c4ab

do not save model if no training

thomwolf committed 6 years ago

466ed991

thomwolf committed 6 years ago

ba910fd8

add do_benchmark

thomwolf committed 6 years ago

59f9cc21

added prediction timing to run_squad and run_classifier

thomwolf committed 6 years ago

3eb54d6d

Typos in readme

thomwolf committed 6 years ago

Verified 786cc412

Merge pull request #119 from danyaljj/patch-1

thomwolf committed 7 years ago

Verified ecc0b54b

Daniel Khashabi committed 7 years ago

Verified 8b1b9394

update readme with information on NVIDIA's apex

thomwolf committed 7 years ago

Verified 8809eb6c

Merge pull request #112 from huggingface/fourth-release

thomwolf committed 7 years ago

Verified e1bfad48

thomwolf committed 7 years ago

d8213588

adding DockerFile

thomwolf committed 7 years ago

37378898

remove logging. basicConfig from library code

thomwolf committed 7 years ago

4a4b0e57

set encoding to 'utf-8' in calls to open

thomwolf committed 7 years ago

ae88eb88

no fp16 on evaluation

thomwolf committed 7 years ago

e1eab59a

fix reloading model for evaluation in examples

thomwolf committed 7 years ago

087798b7

fix swag example for work with apex

thomwolf committed 7 years ago

0f544625

make examples work without apex

thomwolf committed 7 years ago

0cf88ff0

clean up apex integration

thomwolf committed 7 years ago

52c53f39

run_swag example in readme

thomwolf committed 7 years ago

4946c2c5

model loading apex modification

thomwolf committed 7 years ago

d23eed85

include version number + comment in setup.py

thomwolf committed 7 years ago

1cbb32a5

added version in __init__.py

thomwolf committed 7 years ago

ce521776

add saving and loading model in examples

thomwolf committed 7 years ago

d3fcec1a

add pretrained loading from state_dict

thomwolf committed 7 years ago

93f335ef

adding save checkpoint and loading in examples

thomwolf committed 7 years ago

b3caec5a

compatibility PT 1.0 and 0.4.1

thomwolf committed 7 years ago

85fff78c

fixing Adam weights skip in TF convert script

thomwolf committed 7 years ago

13bf0d46

Older