# Shufflenet-v2-Pytorch

**Repository Path**: JiangChenrui/Shufflenet-v2-Pytorch

## Basic Information

- **Project Name**: Shufflenet-v2-Pytorch
- **Description**: Perfect implement. Model shared. x0.5 (Top1:60.646) and 1.0x (Top1:69.402). 
- **Primary Language**: Unknown
- **License**: BSD-2-Clause
- **Default Branch**: master
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2020-08-02
- **Last Updated**: 2020-12-19

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

Shufflenet-v2-Pytorch 
Introduction

This is a Pytorch implementation of faceplusplus's ShuffleNet-v2. For details, please read the following papers: 
	ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design

Pretrained Models on ImageNet

We provide pretrained ShuffleNet-v2 models on ImageNet,which achieve slightly better accuracy rates than the original ones reported in the paper.

The top-1/5 accuracy rates by using single center crop (crop size: 224x224, image size: 256xN): 
Network 		Top-1 	Top-5	Top-1(reported in the paper) 
ShuffleNet-v2-x0.5 	60.646 	81.696	60.300
ShuffleNet-v2-x1 	69.402 	88.374	69.400


Evaluate Models 
python eval.py -a shufflenetv2 --width_mult=0.5 --evaluate=./shufflenetv2_x0.5_60.646_81.696.pth.tar ./ILSVRC2012/
python eval.py -a shufflenetv2 --width_mult=1.0 --evaluate=./shufflenetv2_x1_69.390_88.412.pth.tar ./ILSVRC2012/

Version:
Python2.7
torch0.3.1
torchvision0.2.1

Dataset prepare Refer to https://github.com/facebook/fb.resnet.torch/blob/master/INSTALL.md#download-the-imagenet-dataset