search

rllab-snu / Stage-Wise-CMORL

This is an official GitHub Repository for paper "Stage-Wise Reward Shaping for Acrobatic Robots: A Constrained Multi-Objective Reinforcement Learning Approach".
MIT License
67 stars 8 forks source link

no h1_twohands.yaml and h1_backflip.yaml under algos/student folder when train main_student #1

Closed dbdxnuliba closed 1 month ago

dbdxnuliba commented 1 month ago

no h1_twohands.yaml and h1_backflip.yaml under algos/student folder when train main_student

dbdxnuliba commented 1 month ago

@dobro12 ,hello, could you please provide these two files h1_twohands.yaml and h1_backflip.yaml under algos/student folder ,

dobro12 commented 1 month ago

Hello dbdxnuliba,

Actually, we didn't have the real H1 robot, so we only trained teacher policies for the H1-Twohand and H1-Backflip.

Thus, we didn't provide those yaml files but it can be easily implemented by copying and pasting the "go1_backflip.yaml" and changing the file name to "h1_backflip.yaml". And also change the parameters accordingly. I attached a file, "h1_backflip.yaml", and "h1_twohands.yaml" also can be created in similar ways.

Thanks. Regards

On Mon, Oct 28, 2024 at 9:32 PM dbdxnuliba @.***> wrote:

@dobro12 https://github.com/dobro12 ,hello, could you please provide these two files h1_twohands.yaml and h1_backflip.yaml under algos/student folder ,

— Reply to this email directly, view it on GitHub https://github.com/rllab-snu/Stage-Wise-CMORL/issues/1#issuecomment-2441460918, or unsubscribe https://github.com/notifications/unsubscribe-auth/AHUQNRGNPIMFSUZMVCSTDJLZ5YVEHAVCNFSM6AAAAABP3TWMG6VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDINBRGQ3DAOJRHA . You are receiving this because you were mentioned.Message ID: @.***>

dbdxnuliba commented 1 month ago

ok，I see ，thanks a lot, I will have a try to train h1 student