Calling previous iterations values from an matrix

when I run the code, "index exceeds matrix dimensions" error occurs. My aim is to find maximum row sum and when difference between max value of earlier iteration and current iteration are close (defined by tolerance) then iteration should stop. I am not able to locate error and yet thinking how to code.Can you please tell me the correct code.

for k=1:10 % Main iteration Loop
   for i=1:5 % For matrix creation
    for j=1:5
     A(i,j)=rand % Matrix
    end
   end

     b=zeros(1, 5)
     for i=1:5
      b=b+A(i,:)  % Row sum
     end

     [b_max, index]=max(b)  % find and fprintf of maximum row sum, find its row number and its elements
      if b_max(i)-b_max(i-1)<=0.1  % is it a correct way to call previous iteration values
       fprintf('iterations taken=%d',k)
       fprintf('max value=%14.4f of row% with elements (x1,..,5)=%4.4f',b_max(iteration), index(iteration), A(k,:))
      end

    end

[Joshua]

mukund,

Your error is caused because b_max only has one value in it and you are trying to access b_max(6). I took at stab at cleaning up your code a bit, but I still am a bit confused about what the second fprintf line is trying to accomplish. You never defined the variable iteration and I am not sure what your intent was with index(iteration) because to my knowledge index is not a MATLAB command.

Also, be careful when using this many for loops. If you have 'for i=1:10', i will only change values inside that for loop. After the loop is over, i will keep its last value of 10 for all subsequent calculations.

iteration=10;
for k=1:iteration % Main iteration Loop
    for i=1:5 % For matrix creation
        for j=1:5
            A(i,j)=rand; % Matrix
        end
    end
    b=sum(A);
    b_max(k)=max(b);  % find and fprintf of maximum row sum, find its row number and its elements

end

for m=1:length(b_max)
    for n=1:length(b_max)
        if abs(b_max(m)-b_max(n))<=10  % is it a correct way to call previous iteration values
            fprintf('iterations taken=%d\n',k)
            %fprintf('max value=%14.4f of row% with elements (x1,..,5)=%4.4f',b_max(iteration), index(iteration), A(k,:))
        end
    end
end

Let me know if that helps.

[Mukund]

Thanks Joshua, for correcting my code on b_max and iteration.Through second fprintf command, obtain maximum row sum, the row number that gives maximum sum, and the elements in that row. 1. Obtain matrix A by random numbers, for predefined no. of iterations 2. Obtain row sum of elements 3. Check if max row sum in consecutive iterations are closer, say 0.01. 4. Display iteration no. required, max value of row sum, row number and elements in that row.

[Joshua]

mukund,

Is this maybe what you are looking for? The loop now compares consecutive iterations to each other. Also, I added a third dimension to A so you could store A for each iteration. I still don't think the text output is quite what you want though.

iteration=10;
for k=1:iteration % Main iteration Loop
    for i=1:5 % For matrix creation
        for j=1:5
            A(i,j,k)=rand; % Matrix
        end
    end
    b=sum(A(:,:,k));
    b_max(k)=max(b);  % find and fprintf of maximum row sum, find its row number and its elements
end

for m=2:length(b_max)
          if abs(b_max(m)-b_max(m-1))<=10 
              fprintf('iterations taken=%d\n',iteration)
              fprintf('max value=%14.4f of row\n% with elements (x1,..,5)=%4.4f',b_max(m), m, A(:,:,m))
          end
  end