🕹️ A Minimal Reinforcement Learning Environment Framework for Python . Built by Vishnu Ajit plus ChatGPT. Reduces a huge amount of boilerplate coding when building rl projects.
Uses the pygame library to help visualize the graphical side of things. An advantage it has over the Sam Altman's openaigym is that - openaigym forces you to see a text grid on a terminal output. but this library has better visual interpretation as animations using a pygame window. Try building a snake rl with this. The visual feel is better. Though at the end of the day you should remember that comparing this graphical engine with Sam Altman's openai gym would have the same tradeoff when comparing a Hero Honda Splendor to a Bajaj Pulsar. One engine gives better mileage, one gives better performance. Likewise this library does a brilliant job of showing what the AI is thinking 'visually' simultaneously while it calculates the epsilon values and reward values.
simple_rl_env is a lightweight Python framework for building 2D Reinforcement Learning environments — perfect for beginners learning RL concepts or developers prototyping their own games and agents.
Built from scratch with Python + Pygame, it provides:
- 📦 Simple environment management
- 🎨 Built-in grid rendering system
- 🐍 Example environments (Snake RL)
- ⚙️ Easy hooks for RL agents (DQN, Q-Learning, etc.)
You can install directly from GitHub (until it’s on PyPI):
pip install git+https://github.com/ruforavishnu/simple_rl_env.git