FlappyLearning

FlappyLearning: Open Source ML Framework for Flappy Bird Agents

A free and open source machine learning framework for building flappy bird agents and environments, perfect for reinforcement learning enthusiasts.

What is FlappyLearning?

FlappyLearning is an open-source machine learning framework focused specifically on creating flappy bird agents using reinforcement learning algorithms. It provides an easy-to-use interface and toolkit for developers and hobbyists to build flappy bird environments and train intelligent bots that can play the popular mobile game Flappy Bird autonomously.

The FlappyLearning framework handles all the complexity of setting up the flappy bird environment, simulation engine, and neural network architectures required for training reinforcement learning agents. Users can quickly customize gym environments, experiment with various state representations, action spaces, and reward functions to develop agents with different styles of play.

Out of the box, FlappyLearning offers implementations of popular deep reinforcement learning algorithms like Deep Q-Networks, Policy Gradients, Actor-Critic methods etc. Developers can leverage these reference models or plug in their own custom neural net architectures. The modular design makes it simple to swap different components and tune hyperparameters.

FlappyLearning accelerates research and reduces the barrier to entry for using cutting-edge deep reinforcement learning techniques for this fun and challenging control task. Its simple API allows both experienced machine learning engineers as well as hobbyists new to the field to quickly build and iterate on intelligent flappy bird agents in Python.

FlappyLearning Features

Official Links