triangulate error [message #92826] |
Mon, 07 March 2016 11:26  |
Brian McNoldy
Messages: 35 Registered: July 2000
|
Member |
|
|
I have been battling this for some time now, and can't seem to figure out a solution. I'm cycling through model output times, and the code works fine for the first six or so times, then abruptly ends on one of them, although it doesn't look different, doesn't have NaN's, etc.
Anyway, the error is:
TRIANGULATE: Infinite or invalid (NaN) operands not allowed.
which comes after a simple:
triangulate,xgrid,ygrid,tri
xgrid and ygrid are both 349x339 arrays, and values range from approximately -350 to +350 for each. They are irregularly-spaced, and the difference between adjacent elements is on the order of 2 from each other (e.g. 262.0, 264.1, 266.2).
When I look around for help on this issue, virtually nothing comes up in searches, so I don't know if this has been a common issue, or where I should start looking.
I've tried changing the tolerance in triangulate, defining xgrid & ygrid as floats or doubles, adding tiny random noise to the values... no luck yet.
I'm using IDL 8.4 on 64-bit Linux.
Thank you,
Brian
|
|
|
Re: triangulate error [message #92827 is a reply to message #92826] |
Mon, 07 March 2016 11:47  |
Brian McNoldy
Messages: 35 Registered: July 2000
|
Member |
|
|
I don't know if this helps at all, but when I try:
qhull,xgrid,ygrid,tri,/delaunay
on the same data, I get a more descriptive error, though I still don't understand it.
QHULL: qh_findbest: point p-1 starting at f9 bestoutside? 0 newfacets 1
Brian
|
|
|