xjturobocon

here is a demo of location task. I try to train heading, location and strike, the common problem is the "locomotion" skill of these task, just like the demo. https://user-images.githubusercontent.com/18169613/220001780-c8290b8d-4816-4c11-9bf7-7bf2ebae1a3b.mp4

> In my case, higher disc learning rate while lower disc update frequency might help, in addition to balance reward weights. Can you show me a copy of your setting?...

> Maybe `dataset_strike_amp.yaml` doesn't contain enough locomotion clips? Replacing it by `dataset_reallusion_sword_shield.yaml` seems work well. I also used all motions to train amp location task, but it also didn't work....

>  Can you give me a contact information(WeChat or something else)? I want to ask you for specific information. Thank you.

Hello, I face the same problem as above. It will be very helpful of you to provide a download link to the dataset. Thanks.

> Yes, the latents are sampled randomly during pre-training. Because of the objective used during pre-training, the model will learn to assign different behaviors to different latents automatically. This is...

sorry i didn't ask clearly. What confused me is that after pre-training, is the latent space structured？Is it like the first picture or the second picture below? different color represents...

So how does it have the structure? In pre-training, every step, a latent code is sampled randomly, so there is no guarantee that similar behaviors correspond to similar latent codes,...