Extract some value from genetic algorithm tool addition to fitness function value

Hi,

I use the genetic algorithm-ANN codes for classification. I share you part of my codes.

h= @(x)mse_test(x,net, P, Output);

ga_opt = gaoptimset('display','iter', 'TolFun', 1e-8, 'CrossoverFraction', cr, ...);

[x_ga_opt, final_err_ga, exfl, ouput] = ga(h,10*n+1,ga_opt);

I only can get the final_err_ga value. Bu also confusion matrix(conf) is calculated in ga tool but I cant see it in result screen end of the run. How can I extract the conf variable?

function cal = mse_test(x,net, P, Output)

....

cal=(1/Accuracy);

conf = confusionmat (Output, z);

end

I can share screenshot if necessary. Thank you very much.

Answer 1

[Alan Weiss]

I think that the easiest way would be to run everything in a nested function, as in this example.

Alan Weiss

MATLAB mathematical toolbox documentation

[Ebru PEKEL]

Thank you. But even in this example, matlab gave only the fval value. I want to fval value but not only. I want other values for that result.

I want to believe there is a solution for my problem.

[Alan Weiss]

The nested function allows access to all variable values that you want. Simply pull the values into the main function from the nested functions. See Nested Functions.

Alan Weiss

MATLAB mathematical toolbox documentation

Answer 2

[Ebru PEKEL]

Hi again,

I couldn't get through my problem. Cause I'm not competent at using Matlab I suppose. Could you help me if I upload my all codes in here? My problem is to exract last conf (variable) value from mse_test function.

I would be very grateful if you help. My files are included in the appendix.