How to read a matrix line by line in a for loop from a txt file and save it as a 4-dimensional matrix?

I have a txt file, 'matrix.txt', which has 2 matrices of 1400 rows and 1 column each (they are separeted by a space). To save the file into a matrix, I did the following:

matrix = readmatrix('matrix.txt');

which is 1400x2.

But, I have to read each line and each column in a for loop to store the values into another matrix of dimension 4. The loop is described below:

for i = 1:10
    for j = 1:10
        for k = 1:7
            for a = 1:2
            big_matrix1(i, j, k, a) = matrix( ?, 1)
            big_matrix2(i, j, k, a) = matrix( ?, 2)
            end
        end
    end
end

Then, my question is where I put the question mark. The loop must work in a sequential way, reading each line of matrix( , 1) at a time. E.g.,

big_matrix1(1,1,1,1) = matrix(1,1)
big_matrix1(1,1,1,2) = matrix(2,1)
big_matrix1(1,1,2,1) = matrix(3,1)
big_matrix1(1,1,2,2) = matrix(4,1)
big_matrix1(1,1,3,1) = matrix(5,1)
big_matrix1(1,1,3,2) = matrix(6,1)
etc...

And that is why my matrix() has 1400 rows, because each row (of each column) is the output of each combination of i, j, k and a (10*10*7*2 = 1400).

So, I don't know how to read each line of matrix( , 1) and matrix( , 2) in a sequential way inside my loop. I cannot add another loop for matrix(n,1), where n=1:1400, because it would get me wrong results.

Also, I tried to use fgetl when reading my matrix.txt, by it stores it as a character vector, and I could not figure out how to read 1 column at a time in fgetl.

Thanks in advance for any help!

Answer 1

[Adam Danz]

No need to use loops!

big_matrix1 = permute(reshape(matrix(:,1), [2 7 10 10]), [4,3,2,1]); 
big_matrix2 = permute(reshape(matrix(:,2), [2 7 10 10]), [4,3,2,1]); 

To confirm that the results are the same as your loop method,

c = 0;                         % row counter
big_mat1 = nan(10, 10, 7, 2);  % pre allocation
big_mat2 = big_mat1;           % pre allocation
for i = 1:10
    for j = 1:10
        for k = 1:7
            for a = 1:2
                c = c+1;  % next row (this is a quick hack, for demo purposes)
                big_mat1(i, j, k, a) = matrix( c, 1);
                big_mat2(i, j, k, a) = matrix( c, 2);
            end
        end
    end
end

% Test that the arrays match
isequal(big_matrix1, big_mat1)  %  True
isequal(big_matrix2, big_mat2)  %  True

[Rafaela Rizzo]

Thank you, Adam!