denisyarats/exorl

ExORL: Exploratory Data for Offline Reinforcement Learning

36
/ 100
Emerging

This project helps robotics researchers and practitioners evaluate and improve offline reinforcement learning algorithms. It provides a set of pre-recorded exploratory datasets from various simulated robotic environments and allows you to test how different offline reinforcement learning methods perform on them. The output helps you understand which algorithms are most effective given specific exploratory data.

129 stars. No commits in the last 6 months.

Use this if you are developing or testing offline reinforcement learning models for robotics and want to evaluate their performance against diverse, pre-collected exploratory datasets without needing to run real-world or extensive simulation data collection.

Not ideal if you are looking for a general-purpose reinforcement learning framework for online training or if your primary interest is in designing new data collection strategies rather than evaluating existing datasets.

robotics reinforcement-learning algorithm-evaluation simulation-data offline-learning
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 10 / 25

How are scores calculated?

Stars

129

Forks

9

Language

Python

License

MIT

Last pushed

Feb 08, 2022

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/denisyarats/exorl"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.