Skip to content

Latest commit

 

History

History
24 lines (19 loc) · 3.41 KB

README.md

File metadata and controls

24 lines (19 loc) · 3.41 KB

Mixed Precision Training

Introduction

@article{micikevicius2017mixed,
  title={Mixed precision training},
  author={Micikevicius, Paulius and Narang, Sharan and Alben, Jonah and Diamos, Gregory and Elsen, Erich and Garcia, David and Ginsburg, Boris and Houston, Michael and Kuchaiev, Oleksii and Venkatesh, Ganesh and others},
  journal={arXiv preprint arXiv:1710.03740},
  year={2017}
}

Results and Models

Architecture Backbone Style Conv Lr schd Mem (GB) Inf time (fps) box AP mask AP Config Download
Faster R-CNN R-50 pytorch - 1x 3.4 28.8 37.5 - config model | log
Mask R-CNN R-50 pytorch - 1x 3.6 24.1 38.1 34.7 config model | log
Mask R-CNN R-50 pytorch dconv(c3-c5) 1x 3.0 41.9 37.5 config model | log
Mask R-CNN R-50 pytorch mdconv(c3-c5) 1x 3.1 42.0 37.6 config model | log
Retinanet R-50 pytorch - 1x 2.8 31.6 36.4 config model | log