CuPy adaptation IDW

[Josh von Nonn]

Does anyone have any thoughts on how to adapt this Numpy & Scipy code? Any help would be awesome, thanks!

def idw(data):

 

    valid = np.argwhere(~np.isnan(data))

    tree = spatial.cKDTree(valid)

   

    nans = np.argwhere(np.isnan(data))    

    for row in nans:

        d, idx = tree.query(row, k=12)

        d = np.power(d, -2)

        v = data[valid[idx, 0], valid[idx, 1]]

        data[row[0], row[1]] = np.inner(v, d)/np.sum(d)

       

    return data