913 Bytes

Deep Quaternionary Convolutional Neural Networks

This repository contains code which reproduces experiments presented in the paper ...


Install requirements for experiments with pip:

pip install numpy tensorflow-gpu keras

Depending on your Python installation you might want to use anaconda or other tools.


python install


  1. Get help:

    python scripts/ train --help
  2. Run models:

    python scripts/ train -w WORKDIR --model {real,complex,quaternion} --seg{chiheb,parcollet} --sf STARTFILTER --nb NUMBEROFBLOCKSPERSTAGE

    Other arguments may be added as well; Refer to train --help for

  - Optimizer settings
  - Dropout rate
  - Clipping
  - ...


Please cite our work as