A simple from scratch implementation of Feed Forward Neural Network on the famous Flappy Bird game. The network teaches itself how to play the game through simultaneous generations over a period of time. The framework used is NEAT in Python to teach the AI.
- download the repo.
- open up the directory and run pip install -r requirements.txt.
- now run the game using python FlappyBirdGame.py
- You can configure the Neural Net properties from the config-feedforward.txt file.(eg:- No. of neurons,population etc)
A Huge Shoutout to Tech with Tim, without whom I wouldn't have learnt how to design and make the game from scratch along with the AI behind it.
Link to the playlist if you want to try this on your own: https://www.youtube.com/playlist?list=PLzMcBGfZo4-lwGZWXz5Qgta_YNX3_vLS2