search

alex07143 / Geometric-Robot-DynID

25 stars 10 forks source link

Values for Errors / Estimations #1

Closed AlCap23 closed 4 years ago

AlCap23 commented 4 years ago

Hi!

I was just wondering if it would be possible to provide errors (or estimates if it is allowed ) as either comments in the main.m or as a seperate .mat file. As I understand it, the data in Train_data_human.mat is just a subset of all the data and since I have limited Matlab access, this would be awesome. I am trying to reimplement the algorithms and a benchmark on the original data would be awesome.

Thanks!

alex07143 commented 4 years ago

Hi,

Sorry for the late reply. I uploaded the result file (all the variables in workspace saved as .mat) and figure (.png) that one should get after running "main.m" : 592e64c5a92ffd4be588d17b5d78bc421407089a

Thanks for your interest in our work. Best wishes,

Taeyoon.

AlCap23 commented 4 years ago

Hi!

Thanks a lot! I was on vaccation for the last couple of weeks, so I did not get to use it right now.:)

Best Regards, Julius

shivesh1210 commented 2 years ago

@alex07143 would it be possible for you to share the data of the other two examples (MIT Cheetah and AMBIDEX arm) in the paper?

pwensing commented 2 years ago

@shivesh1210, I have shared the cheetah example over on my lab GitHub (link). It also shows how you can incorporate the entropic regularization using a vector space parameterization of inertia if you need to be using gradient descent rather than SDP.

Please let me know if you have any issues or feedback.