Software Implementation of the Partition of Unity Method

We present a software for efficiently interpolating via kernel bases large datasets in a multivariate setting. It is based on the partition of unity method, and hence on splitting the given (potentially huge) interpolation problem into many small ones. To reach the widest audience among the potential users, we propose both a MATLAB and a Python implementation. We discuss the algorithm details, and hence the paper results in a step-by-step user-friendly tutorial.