Minari
What it is
Minari is a Python library providing a standard format for offline reinforcement learning (RL) datasets. It functions similarly to the Gymnasium or HuggingFace Datasets libraries, offering tools for loading, saving, and managing RL data.
The problem it solves
Lack of a standardized format and utility library for handling offline RL datasets.
What you could use it for
- Conducting offline reinforcement learning research
- Downloading and loading standard RL reference datasets
- Creating and saving custom RL datasets from environment interactions
- Integrating with RL training libraries like TorchRL or d3rlpy
Read the source ↗Back to the library →
Summary written by an AI model from the source; check the source before you rely on it.