Deep_Capsule_Network

Implementation of Deep Capsule Network using Convolutional Dynamic Routing

Introduction

This repository contains the code for our paper titled "Automated Classification of Apoptosis in PhaseContrast Microscopy Using Capsule Network" (link to the paper will be provided after acceptance).

Fig1. Example Deep Capsule Network architecture used in the paper

Load Your Data

Data loaders are provided for MNIST and CIFAR-10 data sets. You may load your data by modifying the provided data loaders or creating your own loader following the similar structure.

Hyper-parameter Tuning

You may change any of the hyper-parameters provided in config.py. For example, to to run the deep capsule network in train mode with batch size of 128, run the following command in terminal:

python main.py --mode=train --model=deep_capsule --batch_size=128