/rl-handson-rlvs21

Stable-Baselines3 (SB3) reinforcement learning tutorial for the Reinforcement Learning Virtual School 2021.

Primary LanguageJupyter NotebookMIT LicenseMIT

Stable Baselines3 Hands-on RL Session

Stable-Baselines3 (SB3) reinforcement learning tutorial for the Reinforcement Learning Virtual School 2021.

Website: https://rl-vs.github.io/rlvs2021/

Stable-Baselines3 repo: https://github.com/DLR-RM/stable-baselines3

SB3 Contrib: https://github.com/Stable-Baselines-Team/stable-baselines3-contrib

Documentation: https://stable-baselines3.readthedocs.io/en/master/

RL Baselines3 zoo: https://github.com/DLR-RM/rl-baselines3-zoo

Content

This Notebook is about getting started with SB3, Gym wrappers, callbacks, multiprocessing and hyperparameter tuning.

Click on the lick to open the notebook in Google Colab: Open Colab Notebook

Solution Notebook

Bonus

Optuna script for automatic hyperparameter tuning.

Acknowledgments

I would like to thank Martiño for helping me improving this hands-on session.