mmsegmentation/configs/fp16
Jerry Jiarui XU 0264de0bb0
[Improvement] Add markdown linter and fix linting errors (#171)
* [Improvement] Add markdown linter and fix linting errors

* fixed pip
2020-10-07 19:50:16 +08:00
..
README.md [Improvement] Add markdown linter and fix linting errors (#171) 2020-10-07 19:50:16 +08:00
deeplabv3_r101-d8_512x1024_80k_fp16_cityscapes.py Support FP16 (#21) 2020-07-20 15:17:18 +08:00
deeplabv3plus_r101-d8_512x1024_80k_fp16_cityscapes.py Support FP16 (#21) 2020-07-20 15:17:18 +08:00
fcn_r101-d8_512x1024_80k_fp16_cityscapes.py Support FP16 (#21) 2020-07-20 15:17:18 +08:00
pspnet_r101-d8_512x1024_80k_fp16_cityscapes.py Support FP16 (#21) 2020-07-20 15:17:18 +08:00

README.md

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

Cityscapes

Method Backbone Crop Size Lr schd Mem (GB) Inf time (fps) mIoU mIoU(ms+flip) download
FCN R-101-D8 512x1024 80000 5.50 2.66 76.80 - model | log
PSPNet R-101-D8 512x1024 80000 5.47 2.68 79.46 - model | log
DeepLabV3 R-101-D8 512x1024 80000 5.91 1.93 80.48 - model | log
DeepLabV3+ R-101-D8 512x1024 80000 6.46 2.60 80.46 - model | log