[SciPy-User] weighted griddata

[Jonathan Stickel]

On 9/16/10 22:41 , scipy-user-request at scipy.org wrote:
>> You may need to fiddle with the smoothing parameter `s=...` for
>> >  SmoothBivariateSpline.
> I was using interpolate.SmoothBivariateSpline previously and had good
> results when it worked, but I struggled with the smoothing parameter
> for so long I had to put it down.  I could not figure out how to
> consistently calculate/estimate/predict a smoothing factor (from the
> data) that was not too low (and never return) or too high (and give
> blocky results).
>
> Is there a way to calculate an optimal smoothing factor from a data set?
>

I know your problem is 2-D, but for 1-D I have written the scikit 
"datasmooth" that implements smoothing by regularization and includes 
generalized cross validation for determining the optimal regularization 
parameter (aka smoothing factor):

http://pypi.python.org/pypi/scikits.datasmooth/0.5

I know the method can be extended to 2-D, but I don't have a need or the 
time to do it myself.

Jonathan