Commits · 7fd54b55a3f7c3134f8cc5a62f4cc447a5cd34de · chenpangpang / transformers

"vscode:/vscode.git/clone" did not exist on "f96413c444a4ce16c6b01770e28a636350df24bf"

03 Dec, 2019 22 commits
- Added support for generic discriminators · 7fd54b55
  piero authored Nov 27, 2019
  
  7fd54b55
- Added a +1 to epoch when saving weights · b0eaff36
  piero authored Nov 27, 2019
  
  b0eaff36
- Added tqdm to preprocessing · 611961ad
  piero authored Nov 27, 2019
  
  611961ad
- Now run_pplm works on cpu. Identical output as before (when using gpu). · afc7dcd9
  piero authored Nov 27, 2019
  
  afc7dcd9
- Cleaned perturb_past. Identical output as before. · 61399e5a
  piero authored Nov 27, 2019
  
  61399e5a
- Fix for making unditioned generation work. Identical output as before. · ffc29354
  piero authored Nov 27, 2019
  
  ffc29354
- Cleaned generate_text_pplm. Identical output as before. · 9f693a0c
  piero authored Nov 27, 2019
  
  9f693a0c
- Renamed SmallConst to SMALL_CONST and introduced BIG_CONST. Identical output as before. · 61a12f79
  piero authored Nov 27, 2019
  
  61a12f79
- Removed commented code. Identical output as before. · ef47b2c0
  piero authored Nov 27, 2019
  
  ef47b2c0
- Removed commented code. Identical output as before. · 7ea12db3
  piero authored Nov 27, 2019
  
  7ea12db3
- Cleaned full_text_generation. Identical output as before. · 08c6e456
  piero authored Nov 27, 2019
  
  08c6e456
- More cleanup for run_model. Identical output as before. · 6c9c1317
  piero authored Nov 27, 2019
  
  6c9c1317
- Improved device specification · 7ffe47c8
  piero authored Nov 27, 2019
  
  7ffe47c8
- First cleanup step, changing function names and passing parameters all the way... · 4f2164e4
  piero authored Nov 27, 2019
```
First cleanup step, changing function names and passing parameters all the way through without using args. Identical output as before.
```
  4f2164e4
- Minor changes · 821de121
  piero authored Nov 27, 2019
  
  821de121
- Fixed minor bug when running training on cuda · 7469d03b
  w4nderlust authored Nov 26, 2019
  
  7469d03b
- Added script for training a discriminator for pplm to use · 0b51fba2
  piero authored Nov 26, 2019
  
  0b51fba2
- Let's make PPLM great again · 34a83faa
  Piero Molino authored Nov 25, 2019
  
  34a83faa
- tokenizer white space: revert to previous behavior · d5faa74c
  Julien Chaumond authored Nov 05, 2019
  
  d5faa74c
- rm extraneous import · 0b77d66a
  Julien Chaumond authored Nov 05, 2019
  
  0b77d66a
- fix the loss backward issue · 83b1e6ac
  Rosanne Liu authored Nov 03, 2019
```
(cherry picked from commit 566468cc984c6ec7e10dfc62b5b4191781a99cd2)
```
  83b1e6ac
- PPLM (squashed) · 572c24cf
  Julien Chaumond authored Nov 05, 2019
```
Co-authored-by: piero <piero@uber.com>
Co-authored-by: Rosanne Liu <mimosavvy@gmail.com>
```
  572c24cf
29 Nov, 2019 2 commits
- Added Camembert to available models · b0ee7c7d
  maxvidal authored Nov 29, 2019
  
  b0ee7c7d
- Refactor logs and fix loss bug · 41aa0e80
  Juha Kiili authored Nov 29, 2019
  
  41aa0e80
27 Nov, 2019 11 commits
- add distilbert + update run_xnli wrt run_glue · d5478b93
  VictorSanh authored Nov 25, 2019
  
  d5478b93
- remove fstrings · 73fe2e73
  VictorSanh authored Nov 05, 2019
  
  73fe2e73
- update readme · 3e7656f7
  VictorSanh authored Nov 05, 2019
  
  3e7656f7
- uniformize w/ the cache_dir update · abd397e9
  VictorSanh authored Nov 05, 2019
  
  abd397e9
- move xnli processor (and utils) to transformers/data/processors · d5910b31
  VictorSanh authored Nov 05, 2019
  
  d5910b31
- change default for XNLI: dev --> test · 289cf4d2
  VictorSanh authored Nov 05, 2019
  
  289cf4d2
- mbert reproducibility results · 84a0b522
  VictorSanh authored Oct 29, 2019
  
  84a0b522
- xnli - output_mode consistency · c4336ecb
  VictorSanh authored Oct 29, 2019
  
  c4336ecb
- add xnli examples/README.md · d52e98ff
  VictorSanh authored Oct 29, 2019
  
  d52e98ff
- run_xnli + utils_xnli · 71f71ddb
  VictorSanh authored Oct 29, 2019
  
  71f71ddb
- Uniformize #1952 · b5d884d2
  Julien Chaumond authored Nov 27, 2019
  
  b5d884d2
26 Nov, 2019 2 commits
- ALBERT for SQuAD · 4374eaea
  Lysandre authored Nov 06, 2019
  
  4374eaea
- Run GLUE and remove LAMB · c110c41f
  Lysandre authored Nov 01, 2019
  
  c110c41f
25 Nov, 2019 3 commits
- Minor bug fixes on run_ner.py · 5d3b8daa
  manansanghi authored Nov 22, 2019
  
  5d3b8daa
- resize model when special tokenizer present · aa92a184
  İbrahim Ethem Demirci authored Nov 16, 2019
  
  aa92a184
- fix #1894 · 7485caef
  Lysandre authored Nov 25, 2019
  
  7485caef