Sum of row and column #?

2 views (last 30 days)
Rachel Dawn
Rachel Dawn on 15 Feb 2018
Answered: Matt J on 15 Feb 2018
Say I ask the user to input the size of a square matrices, n.
And say I want each element of this matrix to be the sum of the row # it's in & the column # it's in. So, for n =2:
2 3
3 4
Is there some Matlab function I could use to calculate each of those elements and display the above as a matrix?

Answers (4)

Matt J
Matt J on 15 Feb 2018
Edited: Matt J on 15 Feb 2018
result=(1:n).'+(1:n);
or for older MATLAB versions,
result = bsxfun(@plus,(1:n).',(1:n));

KSSV
KSSV on 15 Feb 2018
Edited: KSSV on 15 Feb 2018
n = 2 ;
x = 1:n ;
row = repmat(x',1,n) ;
col = repmat(x,n,1) ;
iwant = row+col
Or
x = 1:2*n ;
ind = reshape(x,n,n) ;
[I,J] = ind2sub([n n],ind) ;
iwant = I+J
  1 Comment
Rachel Dawn
Rachel Dawn on 15 Feb 2018
hi! Thank you very much. Could you elaborate on what each line of your code is actually doing/calculating? Sorry, I'm new to Matlab and unfamiliar with all those terms. I tried looking them up but Matlab instructions are quite complicated.

Sign in to comment.


John D'Errico
John D'Errico on 15 Feb 2018
Edited: John D'Errico on 15 Feb 2018
Matt's answer of
result=(1:n).'+(1:n);
is the best way, at least in current versions of MATLAB, because it is most clear what it is doing. But this only applies in 2017 releases and beyond.
Other solutions are less clear, but pretty simple. For example, this should be almost as obvious:
[i,j] = meshgrid(1:n);
result = i + j;
Or
result = cumsum(ones(n),1) + cumsum(ones(n),2);
Or
result = repmat(1:n,n,1) + repmat(1:n,n,1).';
Any of these should be quite clear, and all are equally valid. If unclear still, then it really is time to start reading the getting started tutorials in MATLAB. If you do not follow what they did, then look at each piece of the code. See what it does for some value of n. Read the documentation, perhaps for meshgrid, repmat, bsxfun, cumsum, etc., as appropriate. You won't learn MATLAB until you start reading the documentation.

Matt J
Matt J on 15 Feb 2018
Too bad hankel() isn't MEX-implemented. A MEX-optimized implementation of the following would be O(n),
result = hankel( 2:n+1,n+1:2*n)

Tags

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!