This is a CapsNet repo which is a tenserflow version for Dynamic Routing Between Capsules(Geoffrey E. Hinton et al)

Framework

• Picture below shows the framework of CapsNet, more details can be seen in this repo.

• Picture below shows the reconstruction processing, more details can be seen in this repo.

Routing

Picture below shows how routing work, more details can be seen in this repo.

Requirement

tensorflow 1.3.0

python 2.7.12

numpy 1.13.1

scipy 0.17.0

Usage

(1)download this repo to your own directory

$ git clone https://github.com/nnUyi/CapsNet.git

(2)download mnist dataset and store it in the data directory(directory named data)

(3)training

$ python main.py --is_training=True --mask_with_y=True

(4)testing

$ python main.py --is_training=False --mask_with_y=False

Experiments

The result shows below, after 15 epoches, we can clearly obtain such a good experimental result.

 

Reference

 This repo is finished by referring to naturomics-CapsNet-Tensorflow

Other Implements

• Tensorflow
• Pytorch
• Keras

Contacts

Email:computerscienceyyz@163.com, Thank you for contacting if you find something wrong or if you have some problems!!!