Putting one matrix a varying number of times along the diagonal of another matrix

I have a matrix A which is 6x6. I need to create another matrix where A is along the diagonal with the other entries being 0. I know I could use blkdiag() for this, but the issue is I need to have as many entries of A as the value of another variable that appears earlier in the code. Let's call it n. Sometimes n could be 2, so the matrix would be a 12x12 with A along the diagonal. Sometimes n could be 100 so the ouput matrix would be 600x600 since A needs to appear 100 times. How could I go about automating this process so I don't have to write blkdiag(A,A,A,A...)? Thanks!

Answer 1

[John D'Errico]

A = rand(3);
n = 50;
Acell = repmat({sparse(A)},[1,n]);
B = blkdiag(Acell{:});
spy(B)

Note that I made A sparse in there, so that the final resulting matrix will also be sparse.

[Zain]

This method isn't working for me and I think the issue is with making A sparse. The A I'm working with has some zero entries in itself. Is there a way to do it without making the matrices sparse? I've tried just using A in the same format you used but I get a matrix with all the entries of A as a column vector

[Zain]

I found that this method works well. If n = 4, I could do this and get what I want.

B(1:3,1:3) = A;
B(4:6,4:6) = A;
B(7:9,7:9) = A;
B(10:12,10:12) = A

But I'm having a hard time implementing this iteratively based on n. Any ideas?

[Steven Lord]

M = magic(3)

M = 3×3

8 1 6 3 5 7 4 9 2

N = 4;
C = repmat({M}, 1, N)

C = 1×4 cell array
    {3×3 double}    {3×3 double}    {3×3 double}    {3×3 double}

B = blkdiag(C{:}) % Equivalent to blkdiag(M, M, M, M)

B = 12×12

8 1 6 0 0 0 0 0 0 0 0 0 3 5 7 0 0 0 0 0 0 0 0 0 4 9 2 0 0 0 0 0 0 0 0 0 0 0 0 8 1 6 0 0 0 0 0 0 0 0 0 3 5 7 0 0 0 0 0 0 0 0 0 4 9 2 0 0 0 0 0 0 0 0 0 0 0 0 8 1 6 0 0 0 0 0 0 0 0 0 3 5 7 0 0 0 0 0 0 0 0 0 4 9 2 0 0 0 0 0 0 0 0 0 0 0 0 8 1 6

Search the documentation for "comma-separated list" for an explanation of how that last line works.

[Zain]

Works thank you!!