This repository contains a Reinforcement Learning (RL) project focused on training an AI agent to play the 'Super Mario Bros' game. The project uses the Proximal Policy Optimization (PPO) algorithm and is built on the gym_super_mario_bros environment from OpenAI.
- Environment Setup: Custom environment setup for the Super Mario Bros game.
- Preprocessing: Includes grayscale observation and frame stacking to optimize the learning process.
- Training: Training the RL model using the PPO algorithm with custom callbacks for model saving.
- Testing: Code to test and visualize the performance of the trained model.
Before running the project, ensure you have Python installed on your system. Python 3.7 or later is recommended.
- Clone the Repository
git clone https://github.com/prateekchhikara/supermario cd supermario - Install Dependencies
pip install -r requirements.txt
- Training the Model
python train_rl_model.py
- Testing the Model
python test_model.py
mario_env_setup.py: Setup of the Mario game environment.
preprocess_environment.py: Preprocessing steps for the environment.
train_rl_model.py: Script for training the RL model using PPO.
test_model.py: Script to test and demonstrate the trained model.
callbacks.py: Custom callback definitions for training.
requirements.txt: List of dependencies for the project.
README.md: This file, containing project information and instructions.