DLR-RM / rl-baselines3-zoo

A training framework for Stable Baselines3 reinforcement learning agents, with hyperparameter optimization and pre-trained agents included.
https://rl-baselines3-zoo.readthedocs.io
MIT License
2k stars 510 forks source link

Update TD3/DDPG defaults and upgrade to MuJoCo v4 envs #430

Closed araffin closed 8 months ago

araffin commented 9 months ago

Description

WIP report is here: https://wandb.ai/openrlbenchmark/sbx/reports/SBX-TD3-RL-Zoo-v2-3-0a0-vs-SB3-TD3-RL-Zoo-2-2-1---Vmlldzo2MjUyNTQx

Similar to https://github.com/DLR-RM/stable-baselines3/pull/1785

Motivation and Context

Types of changes

Checklist:

Note: we are using a maximum length of 127 characters per line