search

qgallouedec / panda-gym

Set of robotic environments based on PyBullet physics engine and gymnasium.
MIT License
506 stars 109 forks source link

Are there any benchmarks about dense reward envs? #28

Closed RayYoh closed 2 years ago

RayYoh commented 2 years ago

I am very interesting in the dense reward envs, but I didn't find any benchmarks about them. So, can you provide some benchmarks?

qgallouedec commented 2 years ago

Indeed, there isn't benchmark for dense reward. However, you can easily make one using rl-baselines3-zoo

After installing, run

python train.py --env PandaReach-v1 --algo tqc --env-kwargs reward_type:"'dense'"

Note The current version of rl-zoo only support panda-gym v1 (with is not very different from v2). If you really want to use the v2, please use this branch or wait for this PR to be merged.