Spiking-Neural-Network (SNN)

This project is an implementation of a basic SNN library that supports simple image classification example.

Built With

• scikit-learn
• Numpy
• Matplotlib
• OpenCV

Getting Started

Prerequisites

• Python version 3.11.3 >

  https://www.python.org/

• Virtualenv

  https://pypi.org/project/virtualenv/

Installation

1. Clone the repository

   git clone https://github.com/Albert-Osei/SNN-implementation.git

2. Create Virtual Environment

   virtualenv .spike-venv

   OR

   python -m venv spike-venv

3. Activate Virtual Environment

   source spike-venv\scripts\activate ~ on~windows

   source spike-venv/bin/. activate ~ on~mac

4. Install Dependencies

   pip install -r requirements.txt

Running

• cd into the root where the project is cloned
• run classification/classify.py
• this command will use pregenerated weights and output the results of classifying the test images

Weight Reconstruction (Training)

• to reconstruct the weights, run binary_class/learning.py
• this command will generate new weights file and output neuron[1-3].png which will show the reconstructed weights. One neuron should look random, the other two will produce a pattern similar to the O, and another a pattern similar to the X.