All is correct but this code gives error?

12 views (last 30 days)
Sadiq Akbar
Sadiq Akbar on 19 Nov 2022
Commented: Sadiq Akbar on 21 Nov 2022
I downloaded a file from the Mathworks file exchange and tried to run it, but it was giving error. Then I commented the function definition line and defined what was used ahead. All seems correct, but when I run it, again it gives error. The piece of code is:
clear all
clc
nvars=4;% By Me
fitnessfun=@main3;
%%if nargin<3 | isempty(options)==1
options=hPSOoptions;
%%end
if size(options.space,1)==1
for i=1:nvars
s(i,:)=options.space;
end
options.space=s;
elseif size(options.space,1)~=nvars
error('The rows of options.space are not equal to nvars.');
end
if size(options.maxv,1)==1
for i=1:nvars
v(i,1)=options.maxv;
end
options.maxv=v;
elseif size(options.maxv,1)~=nvars
error('The rows of options.maxv are not equal to nvars.');
end
c1 = options.c1;
c2 = options.c2;
w = options.w;
maxv = options.maxv;
space = options.space;
popul = options.bees;
flights = options.flights;
HybridIter = options.HybridIter;
Show = options.Show;
StallFliLimit = options.StallFliLimit;
StallTimeLimit = options.StallTimeLimit;
TimeLimit = options.TimeLimit;
Goal = options.Goal;
% Define the options for the hybrid approach
options = optimset('LargeScale','off','Display','off','MaxIter',HybridIter);
% Initial population (random start)
ru=rand(popul,size(space,1));
pop=ones(popul,1)*space(:,1)'+ru.*(ones(popul,1)*(space(:,2)-space(:,1))');
% Hill climb of each solution (bee)
for i=1:popul*sign(HybridIter)
[pop(i,:),fxi(i,1)]=fminsearch(fitnessfun,pop(i,:),options,varargin{:});
end
pop=min(pop,ones(popul,1)*space(:,2)');
pop=max(pop,ones(popul,1)*space(:,1)');
fxi=feval(fitnessfun,pop,varargin{:});
Thre required functions are as below:
function e=main3(b)
u=[-30 0 40 60].';% Desired vector
M=5; %1st constant
N=10; %2nd constant
K=3; 3rd constant
d = 0.5; % 5th constant
vec = @(MAT) MAT(:);
vecH = @(MAT) MAT(:).';
steerVecT = @(ang) exp(1j*2*pi*d*(0:M-1).'*sin(vecH(ang))); % Tx steer matrix of size 10 x 3
steerVecR = @(ang) exp(1j*2*pi*d*(0:N-1).'*sin(vecH(ang)));
%%%%%%%%%%%%%%%%%%%%
% Swapping vector b
%%%%%%%%%%%%%%%%%%%%
[~, ix] = sort(u); % u is my desired vector
[~, ix1(ix)] = sort(b);
b = b(ix1);
A = ones(K, 1);
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% Calculation of yo i.e., observed response
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
steerMo = steerMatTR(deg2rad(u), steerVecT, steerVecR);
ro = steerMo*A;
yo = ro;
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% Calculation of ye i.e., estimated y
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
steerMe = steerMatTR(deg2rad(b), steerVecT, steerVecR);
re = steerMe*A;
ye = re;
%%%%%%%%%%%%
% MSE
%%%%%%%%%%%%
abc=0.0;
for m1=1:M*N
abc=abc+(abs(yo(m1,1)-ye(m1,1))).^2;
end
abc=abc/(M*N);
e=abc;
end %%%%%%%%%%% END of our main3s function %%%%%%%%%%%%%%%%%
%%%%%%%%%%%%%%%%%%%%%
% function called inside main3
%%%%%%%%%%%%%%%%%%%%%
function steerM = steerMatTR(targetAngle, steerVecT, steerVecR)
steerA = steerVecT(targetAngle);
steerB = steerVecR(targetAngle);
steerM = zeros(size(steerA, 1)*size(steerB, 1), length(targetAngle));
for idxK = 1 : 1 : length(targetAngle)
steerM(:, idxK) = kron(steerB(:, idxK), steerA(:, idxK));
end
end
After running the above peice of code, it gives the following erro though all seems ok:
Brace indexing into the result of a function call is not supported. Assign the result of
'varargin' to a variable first, then brace index into it.
Error in hPSO1 (line 94)
fxi=feval(fitnessfun,pop,varargin{:});
  2 Comments
Sadiq Akbar
Sadiq Akbar on 19 Nov 2022
Edited: Sadiq Akbar on 20 Nov 2022
Thank you very much dear Jan for your kind response. As I have told in my posted question that the original code was giving an error. So I commented the function definition line i.e.,
" function [x,fval,gfx,output]=hPSO(fitnessfun,nvars,options,varargin)"
and defined the following which were requoired in the forthcoming code:
clear all
clc
nvars=4;% By Me
fitnessfun=@main3;
And then I switched on the debugger and ran it line by line but it gave the following error:
Brace indexing into the result of a function call is not supported. Assign the result of
'varargin' to a variable first, then brace index into it.
Error in hPSO1 (line 94)
fxi=feval(fitnessfun,pop,varargin{:});
Note1: Actually I copied the whole code of hPSO and pasted it in a new editor window and namded it as hPSO1 so that I don't loose the original code of hPSO that's why you see the name hPSO1.
Note2: I took the code from this site:
URL: https://www.mathworks.com/matlabcentral/fileexchange/6497-hybrid-particle-swarm-optimization

Sign in to comment.

Answers (1)

Jan
Jan on 19 Nov 2022
I guess boldly, that the code fails here:
[pop(i,:),fxi(i,1)]=fminsearch(fitnessfun,pop(i,:),options,varargin{:});
...
fxi=feval(fitnessfun,pop,varargin{:});
The posted code seems to be a script. Or did you crop the initial line
function hPSO1
? If so, this was confusing.
The initial clear all will delete all existing variables. On top of a function, this is a crude waste of time. Unfortunately it deletes all loaded functions from the memory also and the reloading is time-consuming also without having any benefits.
What is the meaning of "vargargin{:}" in the two shown lines of code? Is there a variable number of inputs to the shown piece of code (which cannot be a script then)?
By the way, youn simplify
abc=0.0;
for m1=1:M*N
abc=abc+(abs(yo(m1,1)-ye(m1,1))).^2;
end
abc=abc/(M*N);
e=abc;
to
e = mean((yo - ye).^2);
  7 Comments
Sadiq Akbar
Sadiq Akbar on 21 Nov 2022
Thank you very much dear Walter Roberson for your help. I did so and ran it. But it gave me the following error:
File: hPSO.m Line: 3 Column: 41
Invalid expression. Check for missing multiplication operator, missing or unbalanced delimiters,
or other syntax error. To construct matrices, use brackets instead of parentheses.
Error in main (line 3)
[x,fval,gfx,output] = hPSO(@main3, 4, hPSOoptions, varargin{:})

Sign in to comment.

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!