Interpolating on Cubic Grids Using Scipy

[Ali-Tehrani]

Rather than re-inventing the wheel, Scipy has interpn that only works on regular grids (grid axes are orthogonal to coordinate axes) and offers many different kinds of interpolating "linear”, “nearest”, “slinear”, “cubic”, “quintic”, “pchip”, and “splinef2d”. “splinef2d” i"

The following code shows how easy it is using Grid to incorporate inside UniformGrid (assuming grid axes are orthogonal)

    
    x, y, z = grid_cube.get_points_along_axes()
    electro = interpn((x, y, z), electro_cube, kverts, method="linear")#method="cubic", bounds_error=True)

    print("Electro static mean, std, min, max", electro,np.mean(electro), np.std(electro), np.min(electro), np.max(electro))

[shubhamachary56821]

Could You assign this to me, can I work on this issue?

[daksha607]

Hi @PaulWAyers I’m new to open source and would like to work on adding a SciPy-based interpolation path for orthogonal grids. I added a SciPy-based cubic interpolation path using interpn for orthogonal cubic grids when interpolating function values only. The existing CubicSplineimplementation is preserved for derivatives and log-interpolation. I think this will help solve the issue.