Suppose that I would like to optimize a fitness function
fun = @(x) (x(1)) +(x(2)*100);
based on some values for X(1) and X(2)
x(1)=(PSNR4+PSNR3+PSNR2+PSNR1)/4;
x(2)=(1-ratio4)+(1-ratio3)+(1-ratio2)+(1-ratio1)/4;
where all PSNR1 to PSNR4 and ratio1 to ratio4 are real values as seen in Figure.
I have written the following matlab code to do it but it does not work. How can I do it in MATLAB?
fun = @(x) (x(1)) +(x(2)*100);
x = ga(fun,2)
https://www.mathworks.com/help/optim/functionlist.html
fminsearch works, are there constraints (i.e. it is a linear program or quadratic program model, etc)
You can use PSO by Alam
https://www.researchgate.net/publication/297245624_Particle_Swarm_Optimization_Algorithm_and_its_Codes_in_MATLAB
Research Particle Swarm Optimization: Algorithm and its Codes in MATLAB
You may explain more on what does not exactly work on the solution that matlab function provides. You can also try running the same matlab function with some more options for output (just to cross check the solution):
function demoGA
clear all;
[x, fval, reason, output, population, scores] = ga(@ffun2d1, 2, [-1 -1], [1 1])
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
function [f] = ffun2d1(X)
f=20.0+X(1)^2+X(2)^2-10.0*(cos(2.0*pi*X(1))+cos(2.0*pi*X(2)));
end
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
function [f] = ffun2d(X)
f=(1.0/40.0)*(X(1)^2 + X(2)^2)-cos(X(1))-cos(X(2)/sqrt(2.0))+1.0;
end
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
end
PSO could also be an option, but there are several variants of it with possibly different 'power' of speed up to the optimal solution.
- Badges
- Science topic
- Similar topics
- Analytical Chemistry
- Extraction