[Octave-bug-tracker] [bug #53094] findpeaks fails to find peaks in some cases

[Juan Pablo Carbajal]

Follow-up Comment #1, bug #53094 (project octave):

Hi,

I have no clue how findpeaks is implemented in matlab, nor I have that
software to test.

The implementation I provided in octave has several paratmers that you need to
explore before claimiing that the peaks cannot be found. It can perfectly be
that the defaults of octave's findpeaks do not match the behavior of the
default mathwork's findpeaks.

If you can be more precise in your report I might invest time in "fixing" the
functions. Otherwise I would flag this as "works for me"


octave-gui:2> data = [3.7157e+03 3.7494e+03 3.9494e+03 4.3736e+03 5.1826e+03
6.8833e+03 1.2315e+04 5.0602e+04 1.7847e+03 5.4324e+03 4.9405e+04 1.6456e+04
8.1962e+03 5.8239e+03 4.7297e+03 4.1464e+03 3.8383e+03];
octave-gui:3> [i p] = findpeaks(data)
i =

   50602   49405

p =

    8   11


alos make sure that you report bugs agaisnt the development version of the
function which can be found here

https://sourceforge.net/p/octave/signal/ci/default/tree/

    _______________________________________________________

Reply to this item at:

  <http://savannah.gnu.org/bugs/?53094>

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