RLOpt is a flexible and modular framework for Reinforcement Learning (RL) research, built on PyTorch and TorchRL. It is designed to facilitate the implementation, testing, and comparison of various RL agents and optimization techniques. The framework uses dataclass-based configs for library code and Hydra for experiment scripts, allowing convenient customization.
fei-yang-wu/RLOpt
Folders and files
| Name | Name | Last commit date | ||
|---|---|---|---|---|