mohdalibn/flappy-bird-game-ai

⚡An app in which the user can train and test neural networks that are able to play the iconic game flappy bird

This is a Python project featuring a replica of the Flappy Bird game with images sprites found online using a python game library called PyGame. The Neural Network Model was developed using a special genetic algorithm module in python called the Neural Evolution of Augmented Topologies (NEAT) .The design of the mainmenu background, the navigation buttons, and the statistics dashboards were all created using a design prototyping tool called Figma. In this application, the user is able to Train & Test thier model which is stored as a pickle file under the name BestBirdNN.

Install the following Python libraries in your Virtual Environment using PIP.

Note: The library names are CASE-SENSITIVE for PIP installations below. Make sure your type them correctly.

Install PyGame for Python

pip install pygame

Install NEAT for Python

pip install neat-python

Install Numpy for Python

pip install numpy

• Click on the green Code button on the top of this page and download the repository zip using the method of your choice.
• Extract the files into a suitable folder.
• Install all the libraries mentioned above.
• Open the mainmenu.py file on the IDE of your choice. (VSCode Recommended)
• To train your own model, delete the BestBirdNN file first (not needed as the file isn't available in this repository) and then run the mainmenu.py file and click on the Train button on the main menu. Once the training process is complete, a new BestBirdNN file will be created.
• Click on the Test button on the main menu to test your model.