This has scipy as a conditional dependency. We don't do this anywhere in the codebase and at the very least we would need to make it a testing dependency.

I would believe that this interpolation functionality is already provided by the image extension (via agg), depending on scipy is likely a no.

I would believe that this interpolation functionality is already provided by the image extension (via agg), depending on scipy is likely a no.

Agreed - However, I'm not sure this kind of interpolation is even correct for plot_surface which I'd have thought would be more like pcolormesh and either use nearest or Gouraud.

alright, thank you for the feedback, so should we just try to use pcolormesh and either use nearest or Gouraud? Or should we just work on something else since the interpolation is already provided by the image extension via agg?