Using lsqcurvefit with constant parameters

Hi, I am trying to use lsqcurvefit to fit data to an interference function that I created. The problem is that some of the parameters in the function should remain constant through the fitting process (ie. the wavelength of the laser does not change etc...). I have seen some other stabs at this but none that were thorough and actually worked when I ran them. Here is some of my code so you know what I am talking about:

initials = [Aref,Aprobe,w0,z,lambda,phi,x0,y0];
fitfunc = @(initials,x) InterferePhase1(initials,X2);
fitparams = lsqcurvefit(fitfunc,initials, X2 ,imageArray,[], []);

Ideally, I should be able to run this fit and have my initial values for lambda, x0, and y0 stay the same after lsqcurvefit runs.

Let me know if you need any additional information!

Thanks for all your help!

Answer 1

[Matt J]

 initials = [Aref,Aprobe,w0,z,phi];

 fitfunc = @(p,pdata) InterferePhase1([p(1:4),lambda,p(5),x0,y0],pdata);

 fitparams = lsqcurvefit(fitfunc,initials, X2 ,imageArray);

[Chris]

Thanks for your help. I run into problems though with the number of elements because InterferePhase1 requires 8 inputs and now the initials vector has only 5 inputs. It seems to be close though so i'll play around with it.

[Matt J]

As you can see, though, my proposed fitfunc does pass 8 inputs to InterferePhase1 (that was the whole point). It embeds the 5 unknowns in an 8-vector which includes lambda, x0, and y0.

[Chris]

Ya I got it working and understand what you're doing now.

Thanks!