Why am I getting NaN using interp2?

  • I have a 450*90 matrix of values
  • I want to sample the matrix at 100 random sample points
However, while my interp2 function runs without errors, my output results vector contains NaN values after the 90th index.
The code below results in me getting NaN for points after the 90th index of vals - why, and how do I correct it?
Shouldn't I be able to sample 1 point from the matrix, or 50, or 100, or 1000 points?
Code:
row_vect = 1:450;
col_vect = 1:90;
[X,Y] = meshgrid(col_vect, row_vect);
V = abs(rand(450,90));
Xq = row_vect(1:100);
Yq = randi(89, 1,100);
vals = interp2(X, Y, V, Xq', Yq');
Thanks

 Risposta accettata

Torsten
Torsten il 26 Ago 2015

2 voti

You try to extrapolate in Y-direction.
Read the description part for interp2 under
and you'll see why you get NaN for Y-values greater 90.
Best wishes
Torsten.

3 Commenti

Thanks Torsten ....your answer is exactly correct for my question .... however, I made a mistake in my question.
I will edit my original question.
All Y values should be within 1-90.
Use
[X,Y] = meshgrid(row_vect, col_vect);
instead of
[X,Y] = meshgrid(col_vect, row_vect);
Best wishes
Torsten.
I think this worked - thanks a million!!

Accedi per commentare.

Più risposte (1)

Walter Roberson
Walter Roberson il 26 Ago 2015

2 voti

Your problem is your X not your Y. max(X(:)) is 90 because your first argument to meshgrid() is col_vec which is 1:90

Prodotti

Community Treasure Hunt

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

Start Hunting!

Translated by