Flappy AI is a reinforcement learning application for the game Flappy Bird, implemented using the NEAT (NeuroEvolution of Augmenting Topologies) algorithm. The goal of the application is to train a neural network to play Flappy Bird by evolving a population of bird agents over multiple generations.
Ensure you have Python installed on your system.
-
Clone the repository:
git clone https://github.com/your_username/flappy-ai.git cd flappy-ai -
Install dependencies:
pip install pygame neat-python
-
Run the application:
python flappy_ai.py
- The game window displays the Flappy Bird environment, including pipes and a moving base.
- Birds are controlled by neural networks that evolve over generations using the NEAT algorithm.
- The goal is to navigate the birds through the pipes by jumping at the right time.
- The score indicates the number of pipes successfully passed by the birds.
- The game ends when all birds either collide with pipes or reach a score of 100.
The neural network and NEAT algorithm parameters are configured in the config.txt file. You can customize these parameters to experiment with different settings and improve the learning performance.
- The birds are controlled by the neural network and automatically jump based on the learned behavior.
If you want to contribute or modify the game, you can explore the following files:
bird.py: Contains the Bird class, representing the Flappy Bird agents.base.py: Implements the moving base of the game.pipe.py: Defines the Pipe class, representing the pipes that birds must navigate through.flappy_ai.py: The main script that orchestrates the game, NEAT algorithm, and visualization.
- This implementation is based on the NEAT algorithm, pygame library, and the classic Flappy Bird game.
- NEAT (NeuroEvolution of Augmenting Topologies) is a method for evolving artificial neural networks.
- Flappy Bird is a popular mobile game developed by Dong Nguyen.
Enjoy training your AI to master Flappy Bird!