How to train the policy for spinkick?

mingfeisun / DeepMimic_mujoco

DeepMimic Mujoco version; Python only; No Bullet, No C++

103 stars 23 forks source link

How to train the policy for spinkick? #8

Closed MetaAnimation closed 2 years ago

MetaAnimation commented 2 years ago

Thanks for your great work, could you please offer some details about how to design the reward function for the task of spinkick?

mingfeisun commented 2 years ago

The details of reward funciton should be the same as what's been implemented in this codebase. You may need to tune other hyperparameters. You may need to refer to https://github.com/mingfeisun/DeepMimic_mujoco/issues/3 for further details.