How polanarasimharao/gaming works
Project type and purpose
• This is a Python-based AI experiment: using the NEAT (NeuroEvolution of Augmenting Topologies) library to train a neural network that learns to play Flappy Bird.
• There is no separate frontend/backend in the traditional sense—rather, the project combines:
• A game-simulation engine (Flappy Bird via Pygame or similar)
• A NEAT-Python evolutionary algorithm engine
• Model persistence (pickled genomes)
• The goal is to evolve bird agents (neural networks) that maximize fitness (survive as long as possible).
File structure
• Flappy_brd.pk1 – pickled “best” genome (persisted neural network)
• Main.py – orchestrates the training loop: loads config, spawns populations, runs simulation, computes fitness, evolves genomes, saves best model
• Test.py – loads the pickled model and runs an inference-only game demo
• config-feedforward.txt – NEAT configuration (population size, mutation rates, network structure)
• README.md – project description
README for additional insights
• It confirms “Gaming with Python” and NEAT model training.