Pretrained caffe model of ICCV'17 paper:
"ThiNet: A Filter Level Pruning Method for Deep Neural Network Compression"
For more details, please see our project page: [ThiNet Project Page]
Caffe Implementation of ThiNet
224x224 center crop validation accuracy on ImageNet, tested on one M40 GPU with batch_size=32.
| Model | Top-1 | Top-5 | #Param. | #FLOPs | f./b. (ms) |
|---|---|---|---|---|---|
| ThiNet-GAP | 67.34% | 87.92% | 8.32M | 9.34B | 71.73/145.51 |
| ThiNet-Tiny | 59.34% | 81.97% | 1.32M | 2.01B | 29.51/55.83 |
Note: These two models are trained with different image cropping method, see trainval.prototxt for more details.
If you find this work useful for your research, please cite:
@CONFERENCE{ThiNet_ICCV17,
author={Jian-Hao Luo, Jianxin Wu, and Weiyao Lin},
title={ThiNet: A Filter Level Pruning Method for Deep Neural Network Compression},
booktitle={ICCV},
year = {2017},
pages={5058-5066},
}
Feel free to contact me if you have any question (Jian-Hao Luo [email protected] or [email protected]).