Open ChanCheryl opened 5 years ago
hello developers, I want to use my own expert datas, but I don't know how to make expert datas. Could you tell me about how to make? For example: gail's deterministic.trpo.Hopper.0.00.npz
Just load a pre-trained policy, and do env.step(), and save all the states action pairs obtained?
does it matter if the states action pairs are of different lengths for different instances?
hello developers, I want to use my own expert datas, but I don't know how to make expert datas. Could you tell me about how to make? For example: gail's deterministic.trpo.Hopper.0.00.npz