Re: spline with NaNs

[mallet]

James Sherman Jr.-2 wrote:
> 
> Well, as a quick hack, you could do something like:
> 
> valid_indices = ~isnan(param.two);
> P = spline(param.one(valid_indices), param.two(valid_indices));
> 
> In fact, my guess is that this is what Matlab does at the beginning of its
> code, though I don't know for sure.
> 
Hum i’m not sure because in spline.m from Matlab there is:
«% Check that data are acceptable and, if not, try to adjust them
appropriately
[x,y,sizey,endslopes] = chckxy(x,y);»

I don’t know what is chckxy…

Anyway, in my case the NaN was an error of imported data. I fixed it and i
don’t have NaN anymore.

But there still a difference between Octave and Matlab using spline with
NaNs.
-- 
View this message in context: 
http://octave.1599824.n4.nabble.com/spline-with-NaNs-tp2219744p2221412.html
Sent from the Octave - General mailing list archive at Nabble.com.