Commits · 84eb016c5f441d3bbc1acf8466929f74f29d91e7 · OpenDAS / Megatron-LM

25 May, 2021 1 commit
- updating no load rng · fda81a21
  Mostofa Patwary authored May 25, 2021
  
  fda81a21
20 May, 2021 1 commit
- added exit interval for finetuning · d078e54a
  Mostofa Patwary authored May 19, 2021
  
  d078e54a
19 May, 2021 4 commits
- updated the evaluation script for retriver · 825375cf
  Mostofa Patwary authored May 18, 2021
  
  825375cf
- Add temporary assert to finetuning until it can be fixed. · 113c6362
  Jared Casper authored May 18, 2021
  
  113c6362
- renaming the folders · 8e44d619
  Mostofa Patwary authored May 18, 2021
  
  8e44d619
- additional cleaning · 25293807
  Mostofa Patwary authored May 18, 2021
  
  25293807
18 May, 2021 3 commits
- cleaning the code · 2eaf6c79
  Mostofa Patwary authored May 18, 2021
  
  2eaf6c79
- vit pipeline fixes · ccae9dbd
  Vijay Korthikanti authored May 18, 2021
  
  ccae9dbd
- before cleaning the comments · 7a0710ec
  Mostofa Patwary authored May 18, 2021
  
  7a0710ec
13 May, 2021 1 commit
- evaluation works! · f64977fd
  Mostofa Patwary authored May 13, 2021
  
  f64977fd
12 May, 2021 3 commits
- added pre ad post process · 7e335e15
  Mostofa Patwary authored May 12, 2021
  
  7e335e15
- fixing model evaluation of retriver · f9267205
  Mostofa Patwary authored May 12, 2021
  
  f9267205
- DPR finetune and evaluation · 6d03d7af
  Mostofa Patwary authored May 11, 2021
  
  6d03d7af
11 May, 2021 1 commit
- DPR evaluation debugging · 220637f9
  Mostofa Patwary authored May 11, 2021
  
  220637f9
29 Apr, 2021 3 commits
- removed commnets · a8d172b3
  Mostofa Patwary authored Apr 28, 2021
  
  a8d172b3
- removed commnets · f415dc85
  mpatwary authored Apr 28, 2021
  
  f415dc85
- adding dpr code · b9fcb7b4
  mpatwary authored Apr 28, 2021
  
  b9fcb7b4
23 Apr, 2021 1 commit
- implementation dpr · 06076c7a
  Mostofa Patwary authored Apr 23, 2021
  
  06076c7a
09 Apr, 2021 1 commit
- implementing DPR · 10ff0607
  Mostofa Patwary authored Apr 09, 2021
  
  10ff0607
08 Apr, 2021 1 commit
- Release fixes · 50a4b5fa
  Mohammad Shoeybi authored Apr 08, 2021
  
  50a4b5fa
02 Apr, 2021 1 commit
- Addressed MR comments, mostly adding comments to code. · e270f68a
  Jared Casper authored Apr 02, 2021
  
  e270f68a
24 Mar, 2021 2 commits
- Fixing text generation and zeroshot eval and addressing comments. · 64a83fb5
  Jared Casper authored Mar 24, 2021
  
  64a83fb5
- Tasks seems to be working. · b938ec51
  Jared Casper authored Mar 23, 2021
  
  b938ec51
19 Mar, 2021 3 commits
- ICT zeroshot evaluation · fcfd0949
  Mostofa Patwary authored Mar 19, 2021
  
  fcfd0949
- fixed another issue · 40565390
  Mostofa Patwary authored Mar 19, 2021
  
  40565390
- Fixed based on review recoemmendation · 43c9137b
  Mostofa Patwary authored Mar 18, 2021
  
  43c9137b
11 Mar, 2021 1 commit
- made more generic, aligned with other tasks · 661553f6
  Mostofa Patwary authored Mar 11, 2021
  
  661553f6
10 Mar, 2021 1 commit
- ICT zeroshot evaluation code · 8b04e0e4
  Mostofa Patwary authored Mar 10, 2021
  
  8b04e0e4
27 Jan, 2021 2 commits
- added flag so we dont calculate params norm all the time · 3dcbaec9
  mohammad authored Jan 26, 2021
  
  3dcbaec9
- vit model does not get imported automatically · ab507293
  mohammad authored Jan 26, 2021
  
  ab507293
26 Jan, 2021 2 commits
- added params norms · 57d1356e
  mohammad authored Jan 25, 2021
  
  57d1356e
- Adressing more review comments · e6c7b05e
  Vijay Korthikanti authored Jan 25, 2021
  
  e6c7b05e
24 Jan, 2021 1 commit
- added grad norm to logging and tensorboard · 00764415
  mohammad authored Jan 23, 2021
  
  00764415
22 Jan, 2021 1 commit
- Addressing review comments · a7169297
  Vijay Korthikanti authored Jan 22, 2021
  
  a7169297
17 Jan, 2021 1 commit
- GPT2->GPT in zero-shot evaluation scripts · 5ff0f882
  Deepak Narayanan authored Jan 17, 2021
  
  5ff0f882
08 Jan, 2021 1 commit
- vision transformer model and vision classification task · 456f1728
  Vijay Korthikanti authored Jan 08, 2021
  
  456f1728
05 Jan, 2021 3 commits
- Use new api to get loss scale when finetuning. · a13cbe1e
  Jared Casper authored Jan 05, 2021
  
  a13cbe1e
- further refactor, tested, and changed master to main · 983cc311
  mohammad authored Jan 04, 2021
  
  983cc311
- added reload model params for finetuning · 160ba680
  mohammad authored Jan 04, 2021
  
  160ba680
19 Dec, 2020 1 commit
- Quick fix for pipeline tasks to get learning rate correct · 6e83649f
  Jared Casper authored Dec 18, 2020
  
  6e83649f