MichaelFish199 / SonicTheHedgehog2-ReinforcmentLearning Public

Notifications You must be signed in to change notification settings
Fork 0
Star 0

This project implements an agent for playing the SonicTheHedgehog2 game from a ROM file using the Proximal Policy Optimization (PPO) algorithm from the stablebaselines3 library. The agent is trained to learn the optimal actions to take at each step in the game in order to complete the level and maximize the score.

0 stars 0 forks Branches Tags Activity

Name		Name	Last commit message	Last commit date
Latest commit History 6 Commits
Environment_Setup.ipynb		Environment_Setup.ipynb
Model_Parameter_Search.ipynb		Model_Parameter_Search.ipynb
Model_Testing.ipynb		Model_Testing.ipynb
Model_Training.ipynb		Model_Training.ipynb
README.md		README.md

Repository files navigation

SonicTheHedgehog2-ReinforcmentLearning

Creating agent that can play SonicTheHedgehog2 using PPO from stablebaselines3.

This repository contains:

Environment_Setup.ipynb for creating environment
Model_Parameter_Search.ipynb for searching best parameters for model
Model_Training.ipynb for training model
Model_Testing.ipynb for testing model
Results are in ReadMe.md

Sonic_lvl0.mp4

Sonic_lvl1.mp4

About

This project implements an agent for playing the SonicTheHedgehog2 game from a ROM file using the Proximal Policy Optimization (PPO) algorithm from the stablebaselines3 library. The agent is trained to learn the optimal actions to take at each step in the game in order to complete the level and maximize the score.

game reinforcement-learning sonic-the-hedgehog proximal-policy-optimization rom-files stable-baselines3 rewards-and-scoring game-playing-agents

Report repository

Releases

No releases published

Packages

No packages published

Languages

Jupyter Notebook 100.0%