Divide rows of A by first entry of row

I am trying to divide each row of my matrix by its first row entry (to normalize the row).

I can do this for a single row (e.g. row 15 of 2D matrix 'sums'):

sums(15,1:end) ./ sums(15,1)

But this doesn't work:

sums(:,1:end) ./ sums(:,1);

Is there a way to do this in matrix notation, i.e., w/out looping?

Answer 1

[the cyclist]

A_norm = bsxfun(@rdivide,A,A(:,1))

[Matt J]

Minor typo

   sums_norm = bsxfun(@rdivide, sums, sums(:,1));

[Azzi Abdelmalek]

This divide sums by the first column, It's not what the OP asked.

[the cyclist]

I had interpreted his question to mean that he wanted to divide the array by the first row, but after looking more carefully at his code, I agree with you that he wants to divide by the first entry of each row (i.e. the first column).

I have edited my answer accordingly.

[james]

Yes, that works. Your example divides the columns but same idea. I also wanted to divide the rows in place and actually leave the first column unaffected, which is now easily done by:

sums(:,2:end) = bsxfun(@rdivide,sums(:,2:end),sums(:,1))

Thanks for the prompt response. Cheers...James

Answer 2

[Azzi Abdelmalek]

out=A./(repmat(A(1,:),size(A,1),[]))