[Octave-bug-tracker] [bug #53605] (minor) nonzeros.m can be made twice as fast, as follows:

[Rik]

Update of bug #53605 (project octave):

                  Status:                    None => Fixed                  
             Open/Closed:                    Open => Closed                 

    _______________________________________________________

Follow-up Comment #2:

It's a little more complicated than that.  nonzeros(), according to the
documentation, is meant for sparse matrices.  If you try it on sparse matrices
then using the third argument from find() approach is 10X faster than using
indexing.  But for full matrices, the approache you recommended is faster. 
The solution, obviously, is just to check the incoming matrix type and use the
fastest algorithm.


  if (issparse (A))
    [~, ~, v] = find (A);
    v = v(:);
  else
    v = A(find (A));
    v = v(:);
  endif


I checked this in on the development branch of Octave
(http://hg.savannah.gnu.org/hgweb/octave/rev/ff830f8e61ac).


    _______________________________________________________

Reply to this item at:

  <http://savannah.gnu.org/bugs/?53605>

_______________________________________________
  Message sent via/by Savannah
  http://savannah.gnu.org/