how make matrix from a for loop

My code like this,

........... ............

x_error=zeros(1,1);

y_error=zeros(1,1);

z_error=zeros(1,1);

lz=1.5; %lx=1.35; %ly=1.4;

for i=1:1:5 ly=i;

    for j=1:1:5;
       lx=j;

   ......................

    X=abs(x(1,1));
    Y=abs(x(1,2));
    H=abs(x(1,3));

    x_error(i,j)=abs(X-lx);
    y_error(i,j)=abs(Y-ly);
    z_error(i,j)=abs(abs(z1-H)-lz);

    end
end 
x_er=x_error
y_er=y_error
z_er=z_error

it's work but when i change i=1:1:5 to i=1:0.1:5 and j=1:1:5; to j=1:0.1:5;

then error like this >>Attempted to access x_error(1.1,1); index must be a positive integer or logical.

how can i solve this i need to get the x_error value in matrix form not in a cell array form

Answer 1

[Geoff Hayes]

Since the two for loops now iterate with a step size of 0.1, then i and j are no longer positive integer values (1,2,3,4,5, etc.) and so accessing the arrays will fail with the ..index must be a positive integer or logical error message for rational values like 1.1,1.2,1.3,etc.

To accommodate this, you could do the following: realize that the number of elements in 1:0.1:5 is just 41 (try length(1:0.1:5)), pre-allocate memory to your matrices, and use local variables to act as your step sizes:

 n = length(1:0.1:5);

 % pre-allocate memory to the error arrays
 x_error = zeros(n,n);
 y_error = zeros(n,n);
 z_error = zeros(n,n);

 % initialize the step sizes
 lx = 1;
 ly = 1;

 % loop
 for i=1:n

   % re-initialize lx to be one  ---> new code!
   lx = 1;

   for j=1:n
     % do stuff

     % set the errors
     x_error(i,j)=abs(X-lx);
     y_error(i,j)=abs(Y-ly);
     z_error(i,j)=abs(abs(z1-H)-lz);

     % increment lx  ---> new code!
     lx = lx + 0.1;
   end

   % increment ly    ---> new code!
   ly = ly + 0.1;
 end

This way, the code will always be using positive integers to access the error arrays. Try the above and see if it helps!

Answer 2

[Sara]

First, when you allocate variables before the for loop, you need to allocate them so that they can contain all the elements. This means that for the code you have posted, x_error = zeros(5,5) and so on. Second, doing loop where the increment is not an integer is not recommendable for numerical reasons. So, create an array before the for loop with the elements you need to evaluate:

myarray_x = 1:0.1:5;

Then use:

for i = 1:numel(myarray_x)

and inside this loop do:

lx = myarray_x(i);