I am using Least square method in Matlab to solve a Non Linear problem. Problem is bound constrained. When I run the code sometimes I got this error:
"Error using snls (line 47)
Objective function is returning undefined values at initial point. lsqcurvefit cannot continue"
So what can this problem be? I also checked the objective function values at initial points, it is giving negative values.
Can anyone help to figure out this problem? When and how does this occur and how can I solve it?
Thanks in Advance!
Can you share with us an example data set?
We can check it by using 3-4 other programs.
Mahak,
Check your variables. Probably one starts with zero when it is not supposed to. You can initialize to start from somewhere, e.g. at t(1) t may not necessarily equal to zero. I hope I understood you because, your initial question on setting initial values can mean something totally different, e.g. giving initial guess for instance, b0=[2 3]
Dear Demetris & Nicolus, Thank You very much,
Actually I can't share the data due to some administrative privacy.
My objective function is very long (up to 8000 columns in MATLAB editor) with 15-30 unknowns. So I have to set initial values for all variables (or unknowns). Sometime code work but sometimes due to a small change it gives error (written up).
- Badges
- Science topic