I need to analyze data from csv file in Matlab. Since csv file is large Matlab struggles with compilation. What can I do to make it more efficient?

Code reads particular rows from excel inside for loop and puts them into matrix. I need to operate with this matrix but Matlab takes forever to compile it. Here is what I'm currently using:

 n = 0;
 A = zeros(N1, 40);
 B = zeros(N1, 40);
 C = zeros(N1, 40);

 for m = start: N: fin

    n = n + 1;

    section1 = ['B', num2str(m), ':AO', num2str(m)];
    A(n,:) = xlsread(fileName, section1);
    section2 = ['B', num2str(m+1), ':AO', num2str(m+1)];
    B(n,:) = xlsread(fileName, section2);

    C(n,1) = A(n,1) - B(n,1);
    for k = 2:40
        C(n,k) = C(n,k-1) + (A(n,k) - B(n,k));

    end

 end

N1 is large number so that matrices are big. Is there any other efficient way to do this?

Answer 1

[José-Luis]

Ouch!

Don't call xlsread inside a loop. You're always opening the same file, as far as I can see. Just load once and read from that.

[Pirimze Okriashvili]

Thank you for attention! Now that I've removed xlsread from loop there is no problem with any amount of data.