KirbyML (WIP)

A machine learning project for the video game Kirby's Dream Land for the Nintendo Game Boy.

About this Project

The goal of this project is to create a machine learning program that is capable of clearing levels in the video game Kirby's Dream Land for the Nintendo Game Boy.

I decided to pursue this project so I could learn more about machine learning.

Kirby is one of my favorite video game series of all time, and I recently developed an interest in machine learning. I thought this project would be a good starting point and a neat way to include some of my interests.

Requirements

This project requires Python 3.

It is recommended that you have an NVIDIA graphics card and install the CUDA toolkit in order to train the model.

Instructions

1. Clone this repository.

2. Setup a virtual environment

   1. Install virtualenv:

      pip install virtualenv

   2. Create a new Python 3 virtual environment:

      virtualenv venv -p 3

   3. Activate your Python environment

      source venv/bin/activate

   4. Install required Python modules

      pip install -r requirements.txt

3. Obtain a ROM file for Kirby's Dream Land for the Nintendo GameBoy. (The checksum is available in custom_integrations/KirbysDreamLand-GameBoy. You can use this to verify the file.)

4. Run main.py

   python main.py <ROM_PATH> [OPTIONS]
   

Legal Disclaimer

Kirby™ and Kirby's Dream Land™ are properties of Nintendo / HAL Laboratory, Inc..

Nintendo properties are trademarks of Nintendo. © 2018 Nintendo.