This is the implementation of Searching for Accurate Binary Neural Architectures
The implementation of VGG_Small and ResNet18 is in vgg_small.py and resnet18.py
| Network | Expansion Ratio |
|---|---|
| VGG-Auto-A | [0.5, 2, 1, 1, 1, 0.5] |
| VGG-Auto-B | [2, 2, 4, 2, 4, 0.5] |
| Res18-Auto-A | [2, 4, 2, 2, 2, 4, 4, 4, 4, 4, 4, 4] |
| Res18-Auto-B | [3, 4, 4, 3, 1, 4, 4, 4, 3, 3, 3, 3] |
@inproceedings{shen2019searching,
title={Searching for accurate binary neural architectures},
author={Shen, Mingzhu and Han, Kai and Xu, Chunjing and Wang, Yunhe},
booktitle={ICCV Neural Architecture Workshop},
pages={0--0},
year={2019}
}