ModuRL is a modular reinforcement learning (RL) library designed for flexibility and performance. It leverages the candle backend for efficient computations.
-
Modular Design: Easily plug in or extend components.
-
Candle Backend: Efficient and flexible computation using the candle library.
-
Core trait definitions for agents and environments
-
Asynchronous training methods / user-friendly multithreading
-
Potential multi-agent environment support
Contributions are welcome! Please open issues and submit pull requests.
This project is licensed under the MIT License.