Interleaved Group Convolutions for Deep Neural Networks

Introduction

The codes are simple re-implement version for paper: Interleaved Group Convolutions for Deep Neural Networks. (IGCV V1)

Zhang T, Qi G J, Xiao B, et al. Interleaved Group Convolutions for Deep Neural Networks[J]. 2017. arXiv:1707.02725

Some details are different from the description in origin paper.

Structure

We present a simple and modularized neural network architecture, named interleaved group convolutional neural networks (IGCNets). The main point lies in a novel building block, a pair of two successive interleaved group convolutions: primary group convolution and secondary group convolution. The two group convolutions are complementary.

Our motivation comes from the four branch presentation of regular convolution illustrated in the following picture.

Requirements

jupyter notebook
Python3
PyTorch 0.3

Results

We just test IGCV_L24M2 in two datasets: Cifar10 and Tiny ImageNet

Cifar-10

Models	train(Top-1)	validation(Top-1)	L	M	D
IGCV_L24M2	99.4	91.8	24	2	20

Tiny ImageNet

Models	train(Top-1)	validation(Top-1)	L	M	D
IGCV_L24M2	71.2	58.1	24	2	20