This repository provides a collection of Super Mario Bros. environments for reinforcement learning research, offering 32 distinct levels from the original game. It targets RL researchers and developers looking to train agents on classic platforming challenges.

How It Works

The environment bundle leverages the OpenAI Gym interface, allowing seamless integration with standard RL training pipelines. It utilizes a custom game engine or emulator wrapper to render the game states and process player actions, providing observations and rewards for agent training.

Quick Start & Requirements

• Install via pip: pip install gym-pull
• Load environments:

  import gym
  import gym_pull
  gym_pull.pull('github.com/ppaquette/gym-super-mario')
  env = gym.make('ppaquette/SuperMarioBros-1-1-v0')
  

• Requires Python and OpenAI Gym.

Highlighted Details

• Includes 32 levels from the original Super Mario Bros.
• Offers both standard and tile-based observation spaces.
• Compatible with standard Gym API for easy integration.

Maintenance & Community

No specific community channels or maintenance activity are detailed in the README.

Licensing & Compatibility

The license is not specified in the README. Compatibility for commercial use or closed-source linking is undetermined.

Limitations & Caveats

The README does not specify the underlying game engine or emulator used, nor does it detail performance benchmarks or known limitations. The license is also not stated, which may impact commercial adoption.