I have 3D scattered data that I need to interpolate. My chosen interpolation method requires inputs of (x,y,z,v) where:
(x,y,z) spatial coordinates of original grid (1D or 3D arrays) (v) values of data (must be 3D array)
I currently have four 1D arrays for x,y,z, and v with lengths of L=5000 for 5000 unique data points.
Using the solution provided here,
Creating a numpy array of 3D coordinates from three 1D arrays
Numpy.meshgrid produces 3D arrays for each coordinate direction with 1.25 × 10^11 elements. I run out of memory using this approach.
Is there another method to map my 1D array of values, v, to a 3D array with the appropriate size of 5000 elements?
