|
|
Support custom Dictionary implementations in 'preprocess.py' (#448)
|
 Davide Caroselli
|
|
5 years ago |
|
|
Add --input option to interactive.py to support reading from file
|
 Myle Ott
|
|
5 years ago |
|
|
Merge internal changes (#483)
|
Myle Ott
|
|
5 years ago |
|
|
Only use c10d distributed primitives
|
Myle Ott
|
|
5 years ago |
|
|
FIX: '--user-dir' on multi-gpu (#449)
|
Davide Caroselli
|
|
5 years ago |
|
|
Fixed wrong help message shown on '--help' (#446)
|
Davide Caroselli
|
|
5 years ago |
|
|
New command line option '--user-dir' (#440)
|
Davide Caroselli
|
|
5 years ago |
|
|
Merge internal changes (#283)
|
Myle Ott
|
|
5 years ago |
|
|
Merge internal changes (#422)
|
Myle Ott
|
|
5 years ago |
|
|
Add --fp16-scale-tolerance (#397)
|
Myle Ott
|
|
5 years ago |
|
|
Fix arg formatting in preprocess.py and add fmt control for black formatting (#399)
|
Myle Ott
|
|
5 years ago |
|
|
Merge small fixes from internal
|
 Naman Goyal
|
|
5 years ago |
|
|
Merge internal changes
|
Myle Ott
|
|
5 years ago |
|
|
Merge internal changes (#295)
|
Myle Ott
|
|
5 years ago |
|
|
Better support for various c10d API changes
|
Myle Ott
|
|
5 years ago |
|
|
Fix type of c10d bucket size
|
Myle Ott
|
|
5 years ago |
|
|
Add unit test to verify reproducibility after reloading checkpoints
|
Myle Ott
|
|
5 years ago |
|
|
Switch to DistributedDataParallelC10d and bump version 0.5.0 -> 0.6.0
|
 Sergey Edunov
|
|
5 years ago |
|
|
Add documentation
|
Myle Ott
|
|
5 years ago |
|
|
word stats in eval_lm
|
 Alexei Baevski
|
|
5 years ago |
|
|
Diverse Beam Search
|
Myle Ott
|
|
5 years ago |
|
|
add flag that allows keeping optimizer config
|
alexeib
|
|
5 years ago |
|
|
Merge internal changes
|
Myle Ott
|
|
5 years ago |
|
|
option to print language model words and their log probs during evaluation
|
Alexei Baevski
|
|
5 years ago |
|
|
disable printing alignment by default (for perf) and add a flag to enable it
|
Alexei Baevski
|
|
5 years ago |
|
|
add model override argument from load_ensemble_for_inference at generation time, updating readme for stories
|
 Angela Fan
|
|
5 years ago |
|
|
Support FP16 during inference
|
Myle Ott
|
|
6 years ago |
|
|
Better failure message when loss explodes during FP16 training
|
Myle Ott
|
|
6 years ago |
|
|
Change --path to be colon-separated instead of comma-separated
|
Myle Ott
|
|
6 years ago |
|
|
Add FairseqTask
|
Myle Ott
|
|
6 years ago |
Connect to our Discord channel
Models
Annotations