How can I increase the size of matrices in cell arrays with each iteration of a loop?

I have a cell array that is 10x10 with each cell containing a 1x3 matrix of numerical values. This cell array is generated inside a loop that has 10 iterations, thus there are 10 versions of this cell array, each with different values.

I would like to know how to store each iteration of this cell array in a single cell array which is still 10x10, however with each cell containing a 10x3 matrix. Basically, I want each iteration of the loop to add a row to the matrix in each cell and populate those rows with the corresponding cells of the new cell array generated during that iteration.

Answer 1

[Voss]

A 10x10 cell array where each cell contains a 1x3 vector:

C = repmat({[0 0 0]},10,10);

Another 10x10 cell array where each cell contains a 1x3 vector:

C_new = repmat({[1 1 1]},10,10);

Append the contents of each cell of C_new as a new row on the contents of the corresponding cell of C:

C = cellfun(@vertcat,C,C_new,'UniformOutput',false)

C = 10×10 cell array
    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}
    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}
    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}
    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}
    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}
    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}
    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}
    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}
    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}
    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}    {2×3 double}

So that last line is what you would do on each iteration, where C_new is the new version each time.

% Initialize C
C = cell(10,10);

% run 10 iterations
for ii = 1:10
    % ... 
    % here you would calculate C_new, which is the new version for this iteration.
    % ...
    C = cellfun(@vertcat,C,C_new,'UniformOutput',false);
end
disp(C)

    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}
    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}
    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}
    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}
    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}
    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}
    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}
    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}
    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}
    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}    {10×3 double}

[Jacob Martire]

Thanks!

[Voss]

You're welcome!

Answer 2

[Sulaymon Eshkabilov]

There are a few different ways to create such as cell array. If to get it done in a loop:

rng(1)  % To replicate at any time and obtain the same results of random numbers
for ii = 1:10
    for k=1:10
        H{ii,k} = randi([-5,5],1, 3);
    end
end
H{1,1}

ans = 1×3

-1 2 -5

H{1,2}

ans = 1×3

-2 -4 -4

H{3,1}

ans = 1×3

-4 -1 2

% etc

Answer 3

[Fifteen12]

This code gives two solutions, one that can be implemented after your original iteration (it just harvests the values and puts them into a new matrix), and one that can be implemented during your original iteration (which you'll have to splice into your own code).

% Generate original cell arrays
n = 10;

for k = 1:n
    for i = 1:n
        for j = 1:n
           foo{i, j} = [i, j, k]; %Random values
        end
    end
    original{k} = foo;
end

% Post adaptation from original iteration
for i = 1:n
    for j = 1:n
       temp = zeros(n, 3);
       for k = 1:length(original)
           temp(k, :) = original{k}{i, j};
       end
       out1{i, j} = temp;
    end
end

% Adaption during original iteration
out2 = cell(n, n);
for k = 1:n
    for i = 1:n
        for j = 1:n
           out2{i, j}(k, :) = [i, j, k]; %Random values
        end
    end
end

[Jacob Martire]

This works well, thank you very much!