This repository follows after my final year project. The aim of the project was to reduce the memory consumption of a LightGBM model used for malware detection. To accomplish this, I used the Joint Mutual Information Maximisation (JMIM) algorithm to select the
My original implementation has the following issues:
- Not language agnostic i.e., only suitable for Python programs.
- Adopts a slow, naive approach to handle discrete and continuous variables.
The purpose of this repository is to use JMIM to create a tool which selects the
For those interested to know how the algorithm above was derived go to section 3.4.1 of my final year project paper.
For Windows users:
- Clone the repository:
git clone https://github.com/DeraUchenwoke/jmim.git
- In PowerShell terminal run:
cd scripts
followed by.\setup.ps1
.
Google style guide. Powershell practice and style.
The tool was written in line with CPP core guidelines.