Chair Classification by function

This code was tested with python 3.7.

Install dependencies

python -m pip install -r requirements.txt

Train

This script is based on 4 type function of chair dataset as an example. For training, please run:

python TestRun.py

Script Introduction

SE_Block.py is a channel-wise attention that used to select most important feature map by Squeeze-and-Excitation [J. Hu, CVPR'18].

ResNet.py is the Deep Residual Network that use residual learning to solve the vanishing gradient problem at deep neural network [K. He, CVPR'16].

SENet.py follow the sequeeze and excitation, "capture features in the convolution", to make the network more efficient.

SP&A-Net-Test-Run.ipynb is in the form of a Jupyter Notebook as a simple display with chair dataset as the training object.