How Do the "fittype" function works

I am working on a project where I have got a piece of code where the fittype function has been used like this

ft = fittype( 'cconv(A.*exp(-x.*B), y,length(x))' , 'independent', {'x', 'y'}, 'dependent', 'z' );

where the circuler convolution is beeing done with two vectors A and B . My query is where the values of A and B comes from or how the values are beeing initialized .

[Arun Badigannavar]

Please elaborate your question, A, B depends on vector you choose,

[somnath]

Actually ai am studying a piece of code and After the fittype function the "fit" function is called , I want to know how the cconv((A.*exp(-x.*B), y,length(x)) function is called and A and B are estimated.

Answer 1

[Walter Roberson]

fittype() does not initialize A or B in your example. fittype() creates a model that includes an anonymous function that will name A and B as parameters. Usually the fit() function is then used to determine the parameters, so what you want to know is how it initializes the values.

The answer to that is described http://www.mathworks.com/help/curvefit/fit.html#input_argument_namevalue_startpoint

If no start points (the default value of an empty vector) are passed to the fit function, starting points for some library models are determined heuristically. For rational and Weibull models, and all custom nonlinear models, the toolbox selects default initial values for coefficients uniformly at random from the interval (0,1).

Yours is a custom nonlinear model.

[somnath]

Thankyou very much . Now I got how the coeefs are initialized . Can you please help me understand how the circulerconvolution(with the function cconv()) is beeing done.