[Octave-bug-tracker] [bug #57185] eigs does not consistently order eigenvalues

[Rik]

Follow-up Comment #13, bug #57185 (project octave):

This is the behavior we need to match for Matlab compatibility.


Value of sigma   Output sorting
----------------------------------------------

'largestabs'     Descending order by magnitude

'largestreal'    Descending order by real part

'largestimag'    Descending order by imaginary part

'smallestabs'    Ascending order by magnitude

'smallestreal'   Ascending order by real part
'bothendsreal'   Ascending order by real part

'smallestimag'   Ascending order by imaginary part

'bothendsimag'   Descending order by absolute value of imaginary part



    _______________________________________________________

Reply to this item at:

  <https://savannah.gnu.org/bugs/?57185>

_______________________________________________
  Message sent via Savannah
  https://savannah.gnu.org/