Commits · efb44a8310f68a1ce4e5633b123c9bff3456f0bf · chenpangpang / transformers

"benchmark/vscode:/vscode.git/clone" did not exist on "9d61205dac19ce9a15e2f3238b75bfe3d7c913af"

04 Nov, 2018 13 commits
- distributed in extract features · efb44a83
  thomwolf authored Nov 04, 2018
  
  efb44a83
- update float() · d9d7d1a4
  thomwolf authored Nov 04, 2018
  
  d9d7d1a4
- remove old methods · c6207d85
  thomwolf authored Nov 04, 2018
  
  c6207d85
- add distributed training · 965b2565
  thomwolf authored Nov 04, 2018
  
  965b2565
- add gradient accumulation · 1ceac85e
  thomwolf authored Nov 04, 2018
  
  1ceac85e
- clean up · 6b0da96b
  thomwolf authored Nov 04, 2018
  
  6b0da96b
- logging + update copyright · 834b485b
  thomwolf authored Nov 04, 2018
  
  834b485b
- multi-gpu cleanup · 1701291e
  thomwolf authored Nov 04, 2018
  
  1701291e
- what's in loss again · 5ee17168
  thomwolf authored Nov 04, 2018
  
  5ee17168
- add tqdm, clean up logging · 0b7a20c6
  thomwolf authored Nov 04, 2018
  
  0b7a20c6
- add numpy import · d4e3cf35
  thomwolf authored Nov 04, 2018
  
  d4e3cf35
- remove run_squad_pytorch · cf366417
  thomwolf authored Nov 04, 2018
  
  cf366417
- fixing verbose_argument · 26bdef43
  thomwolf authored Nov 04, 2018
  
  26bdef43
03 Nov, 2018 27 commits
- tweaking the readme · d6418c5e
  thomwolf authored Nov 03, 2018
  
  d6418c5e
- update readme · 3b70b270
  thomwolf authored Nov 03, 2018
  
  3b70b270
- Merge branch 'master' of https://github.com/huggingface/pytorch-pretrained-BERT · eaa6db92
  thomwolf authored Nov 03, 2018
  
  eaa6db92
- update readme, file names, removing TF code, moving tests · f8276008
  thomwolf authored Nov 03, 2018
  
  f8276008
- fix typo · f18ae210
  Ubuntu authored Nov 03, 2018
  
  f18ae210
- Multi-Gpu loss - Cleaning · 3c24e4be
  VictorSanh authored Nov 03, 2018
  
  3c24e4be
- WIP modeling_test_pytorch.py · 5de1517d
  Tim Rault authored Nov 03, 2018
  
  5de1517d
- fix typo · 1ba5b58c
  VictorSanh authored Nov 03, 2018
  
  1ba5b58c
- Fix both loss and eval metrics -> more coherence on the loss (eval vs train and tf vs pt) · 5858e8e4
  VictorSanh authored Nov 03, 2018
  
  5858e8e4
- Fix import on initalization · cd09cd5b
  VictorSanh authored Nov 03, 2018
  
  cd09cd5b
- WIP · ec66841a
  Tim Rault authored Nov 03, 2018
  
  ec66841a
- Merge branch 'master' of https://github.com/huggingface/pytorch-pretrained-BERT · 139873f6
  thomwolf authored Nov 03, 2018
  
  139873f6
- special edition script · 04287a4d
  thomwolf authored Nov 03, 2018
  
  04287a4d
- Add seed in initialization · a1af5247
  VictorSanh authored Nov 03, 2018
  
  a1af5247
- Fix loss loss logging for multi-gpu compatibility · 4faeb38b
  Ubuntu authored Nov 03, 2018
  
  4faeb38b
- update optimizer run_squad · 25f73add
  thomwolf authored Nov 03, 2018
  
  25f73add
- update run_squad with tqdm · f514cbbf
  thomwolf authored Nov 03, 2018
  
  f514cbbf
- add model.zero_grad() · cb76c1dd
  thomwolf authored Nov 03, 2018
  
  cb76c1dd
- Merge branch 'master' of https://github.com/huggingface/pytorch-pretrained-BERT · a4086c5d
  thomwolf authored Nov 03, 2018
  
  a4086c5d
- fixing optimization · 088ad458
  thomwolf authored Nov 03, 2018
  
  088ad458
- typo on tokenization · 8bd6b235
  VictorSanh authored Nov 03, 2018
  
  8bd6b235
- `scatter_` and `scatter` · 2c55568c
  VictorSanh authored Nov 03, 2018
  
  2c55568c
- Merge pull request #1 from huggingface/multi-gpu-support · a6efe123
  VictorSanh authored Nov 03, 2018
```
Create DataParallel model if several GPUs
```
  a6efe123
- Create DataParallel model if several GPUs · 5f432480
  VictorSanh authored Nov 03, 2018
  
  5f432480
- Update README.md · 5889765a
  VictorSanh authored Nov 03, 2018
  
  5889765a
- Update the comparison notebook · 8c932e37
  VictorSanh authored Nov 03, 2018
  
  8c932e37
- Small typo in `trange` · 391a4ec2
  VictorSanh authored Nov 03, 2018
```
I seriously don't understand why they defined num_train_epochs as a float in the originial tf code.
I Will change it at the end to avoir merge conflicts for now.
```
  391a4ec2