I have a dataset that conforms to the restriction that if x2 >= x1 and y2 >= y1, then f(x2, y2) >= f(x1, y1). I would like to find a method of interpolation that preserves this assumption even among the interpolated values. Using e.g. scipy.interpolate.LinearNDInterpolator does not, and I suspect it's because (for example) as you move up, you might move closer to a point that is left of you, bringing the interpolated value down. Is there an existing interpolation implementation that will preserve upward-rightward monotonicity? Continue reading...
