Faster repeated interp3 and griddata

Hi,

I am asking a question for interp3 and griddata , but in principle the problem is the same.

Suppose i have N-functions with different data scattered data in 3D which i want to interpolate to a structured 3D grid using griddata.

The locations of the scattered data and the new 3D grid is always the same. Is it possible to accelerate the process for multiple interpolations given that the geometry information is the same for all N-functions?

Repeating the inverse (from 3D structured grid to scattered points ) i am using interp3. I would like to improve speed also in this process, since the geometry is the same too.

Thank you,

Redi

Answer 1

[Steven Lord]

Create a scatteredInterpolant object and change the Values property for each of your new set of data as per the "Replacement of Sample Values" example on that page. For the other direction, gridded data to scattered points, see griddedInterpolant.

[REDI PONI]

Thanks.

Works fine.