Fittype TestAnonymousFunction Argument Error

Can someone please explain why I keep getting this error?

Error using fittype>iTestAnonymousFunctionArgumentOrder (line 811) The coefficients must come before the problem parameters.

I have my function as:

fun=@(F,PS,fp,fis,x)...
    longobjfun(x,F,PS,fp,fis,A);
ft = fittype(...
    fun, ...
    'problem',{'A'},...
    'coeff', {'F','PS','fp','fis'});

I know that for fittype, the anonymous function has to have a form of:

fun = @(fitparameters, problemparameters, independent) expression;

It stated that problemparameters can be optional. How do I remove this? I'm not sure what to do with this.

Answer 1

[Chandani Madnani]

Modify the code as shown below to pass the Problem parameter and see if it works:

fun=@(F,PS,fp,fis,A,x)... longobjfun(x,F,PS,fp,fis,A); ft = fittype(... fun, ... 'problem',{'A'},... 'coeff', {'F','PS','fp','fis'});

[Arbol]

I have tried and rearrange like yours, I get this error: Error using fittype/testCustomModelEvaluation (line 12) Expression model_2cxm_cfit(x,fp,PS,vp,ve,A) is not a valid MATLAB expression, has non-scalar coefficients, or cannot be evaluated: Error while trying to evaluate FITTYPE function :

Matrix dimensions must agree.

[Arbol]

Also, on one note, can A be a vector? the problem value?

[Arbol]

I have got this fixed. But I have no idea how it was fixed. It could due to that i didnt clear the fittype, so it still keep the old fittype and carry over to the next run.

[Arbol]

My fix is as followed (for future reference if someone look at this):

 fun=@(F,PS,fp,fis,x)... 
longobjfun(F,PS,fp,fis,A,x); 
ft = fittype(fun, ... 
'independent',{'x'},... 
'coeff', {'F','PS','fp','fis'});