data-driven-dynamics icon indicating copy to clipboard operation
data-driven-dynamics copied to clipboard

Adapt QP Optimizer for unconstraint parameters

Open manumerous opened this issue 3 years ago • 3 comments

Currently the QP optimizer requires all parameters to be either fixed or bounded. We sould add the option for unconstraint parameters as well.

@Jaeyoung-Lim what do you thing about just not specifying bounds for certain parameters in the config? I guess thats the easiest way.

manumerous avatar Sep 11 '21 17:09 manumerous

@Jaeyoung-Lim what do you thing about just not specifying bounds for certain parameters in the config? I guess thats the easiest way.

Yes, I think that would work great. If you still have to pass somekind of a upper bound to the optimizer, you can pass something link FLOAT32_MAX (not sure how this is defined in python)

Jaeyoung-Lim avatar Sep 11 '21 17:09 Jaeyoung-Lim

I think this is handled by the cvxpy library. So we do not need to worry about that.

manumerous avatar Sep 11 '21 17:09 manumerous

@manumerous Hmm... then why do we need it in the first place?

Jaeyoung-Lim avatar Sep 11 '21 18:09 Jaeyoung-Lim