search

DLR-RM / rl-trained-agents

A collection of pre-trained RL agents using Stable Baselines3
https://huggingface.co/sb3
MIT License
105 stars 25 forks source link

Improved model performance: TD3 in Walker2DBulletEnv-v0 #4

Closed modanesh closed 3 years ago

modanesh commented 3 years ago

The performance of the already provided model was around 194.60 +/- 79.90 on 10 test episodes. I trained a new agent which has a much better performance on 10 test episodes: 2240.81 +/- 6.36.

I used the rl baseline3 zoo code for training.

araffin commented 3 years ago

Hello,

Thanks =) The performance was indeed far away from what it can be.