I had a working python code that generates an interpolating function from row data:
scipy.interpolate.interp1d(Static_V_E[:,0], Static_V_E[:,1], kind = INTERP_KIND, bounds_error=False, fill_value="extrapolate")
It was working with INTERP_KIND = "cubic"
and it worked until a few days ago (I do not remember if I updated scipy/anaconda or something else). However, now the code crashes with the error:
ValueError: Extrapolation does not work with kind=spline
And I have to replace it with INTERP_KIND = "linear"
.
This, of course, leads to much worse results in my calculation. Is there any way to have a better interpolation than the linear one, possibly without changing the syntax? (I have this function called hundred of times in the code, and I which to be able just to change INTERP_KIND or something else).