Skip to content

Latest commit

 

History

History
39 lines (33 loc) · 1.48 KB

README.md

File metadata and controls

39 lines (33 loc) · 1.48 KB
Raw

FlappyBird_AI

Live site

https://flappybird-ai.netlify.app

Screenshot

Screenshot

About

This AI was trained using N.E.A.T. This is a genetic algorithm that trains the AI by creating generations of agents and simulating them in a environment. Each bird has a Neural Network that will decide to make the bird flap based on its inputs. The birds with the best Neural Networks are selected for the next generation and slightly mutated to create new behavior.

This project was made using P5js to render the game and TensorFlow js to create the Neural Networks.

Run Local

Download the code and navigate to this directory.

git clone https://github.com/SeanFitz1997/FlappyBird_AI.git
cd FlappyBird_AI

Run a python server and open the index.html file.

$ python -m http.server
Serving HTTP on 0.0.0.0 port 8000 (http://0.0.0.0:8000/) ...

Whats here

  • data (Contains trained model)
  • images (Contains game and website images)
  • src (Contains code for Flappy Bird AI)
    • main.js (setup and draw loop)
    • Environment.js (render and move environment)
    • Ground.js (render ground)
    • Pipe.js (render and move pipe)
    • Population.js (render and move population of birds)
    • Bird.js (render and make actions using Neural network)
    • Brain.js (create Neural network and make prediction)
    • Neuroevolution.js (evaluate and create next generation)
  • index.html (webpage)
  • index.js (handles webpage interactions)