craigwh10 / java-naive-bayes

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Package for running naive bayes forecasting for gaussian distributed datasets.

Aiming to work against an example of classifying cryptocurrencies using kaggle dataset


To run

  • Requires Java 17 LTS.
  • As defined in build.gradle, to change this in IDE on Mac-IntelliJ, go to file > project structure > project > jdk

Then..

gradle run

Tests:

Uses assertJ.

gradle test

Todo:

  • Stretch tasks
    • Remove lambda expressions
    • Performance monitoring
    • Graphics generation for distribution and time series generation
  • Gaussian distribution
    • Generating classes for classifier
    • Stat methods
    • Naive bayes implementation

Notes:

  • To use a modular setup right click the java dir, open module settings then control them there, otherwise modules are added outside the scope of the core reverse dns style package.

About


Languages

Language:Java 100.0%