MobileNet-V2-Mxnet

Introduction

This is a Mxnet implementation of Google's MobileNets (v2). For details, please read the following papers:

• [v2] Inverted Residuals and Linear Bottlenecks: Mobile Networks for Classification, Detection and Segmentation

This Model is converted from MobileNet-Caffe

Pretrained Models on ImageNet

We provide pretrained MobileNet 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	sha256sum	Architecture
MobileNet v2	71.90	90.49	a3124ce7 (13.5 MB)	netscope

Evaluate Models with a single image

Evaluate MobileNet v2:

python mobile_v2_eval.py

Expected Outputs,the output is different from caffe version, it could be cause by difference in preprocess:

(0.38443729, "'n02124075 Egyptian cat'")
(0.11267297, "'n02123159 tiger cat'")
(0.074108832, "'n02123045 tabby, tabby cat'")
(0.058392685, "'n02119022 red fox, Vulpes vulpes'")
(0.031767886, "'n02094258 Norwich terrier'")

Todo List

• convert symbol
• add python symbol