How can I finetune the policy in real world

[Yu Zhang]

I'm working on a real robot Laikago and I have no problem training the policy in the simulation.

However, I fail to find codes to finetune it in the real world, which is called domain adaptation in paper.

Please let me know if I miss something.

Thanks a lot.

[xbpeng]

Hi,
This repo only contains the code for motion imitation in simulation. We haven't released the code for domain adaptation.

[Yu Zhang]

Thanks for your reply.
Do you have any plan to release it? or you will keep it private?
I think domain adaptation make great contribution to the results, so it would be great if you publish it.

[xbpeng]

Sorry but no plans to release that code at the moment. Feel free to ask if you have any questions though.

[Yu Zhang]

That is unfortunate, but still thanks for your notice.