Python to C++ Project on Separable-CNN in MobileNets Paper.

---

Introduction

we propose what point affect to accuracy from between Python Package and C raw programming without Package.

Requirements

• Python code about Separable-CNN
• C code about Separable-CNN
• visual studio 2015
• python 2.7

Datasets

Train Image Dataset

• download dataset from this link and put it in this project

Test Image Dataset

The test image dataset are sampled from this link and put ti in this project

TODO

• Back propagation comparing between Python Package and C without Package just raw programming

DONE

• Paper
• Python Package programming study
• forward propagation C coding(CNN, LightNormalization, Relu, Pool, Depthwise-CNN, Pointwise-CNN, Affine, Softmax)
• Comparing layer by layer
• inference 1 dataset (cifar10)
• valid dataset
• Get Input Image about Normal airplane

Image Input

** Input size Reduction **

• 788 * 526 size image source

• 32 * 32 size input

Reference

• Andrew G. Howard, Menglong Zhu, Bo Chen, Dmitry Kalenichenko, Weijun Wang, Tobias Weyand, Marco Andreetto, Hartwig Adam. MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications. arXiv:1704.04861, 2017