Question

I’m trying to copy the Scipy Cookbook function:

from scipy import ogrid, sin, mgrid, ndimage, array
x,y = ogrid[-1:1:5j,-1:1:5j]
fvals = sin(x)*sin(y)
newx,newy = mgrid[-1:1:100j,-1:1:100j]
x0 = x[0,0]
y0 = y[0,0]
dx = x[1,0] - x0
dy = y[0,1] - y0
ivals = (newx - x0)/dx
jvals = (newy - y0)/dy
coords = array([ivals, jvals])
newf = ndimage.map_coordinates(fvals, coords)

by using my own function that has to work for many scenarios

import scipy
import numpy as np
"""N-D interpolation for equally-spaced data"""                                         
x = np.c_[plist['modx']]                                                                
y = np.transpose(np.c_[plist['mody']])                                                  
pdb.set_trace()                                                                         
#newx,newy = np.meshgrid(plist['newx'],plist['newy'])                                   
newx,newy = scipy.mgrid[plist['modx'][0]:plist['modx'][-1]:-plist['remapto'],               
                     plist['mody'][0]:plist['mody'][-1]:-plist['remapto']]                                                                                        
x0 = x[0,0]                                                                             
y0 = y[0,0]                                                                             
dx = x[1,0] - x0                                                                        
dy = y[0,1] - y0                                                                        
ivals = (newx - x0)/dx                                                                  
jvals = (newy - y0)/dy                                                                  
coords = scipy.array([ivals, jvals])                                                    
for i in np.arange(ivals.shape[0]):                                                     
    nvals[i] = scipy.ndimage.map_coordinates(ivals[i], coords)                                                                                                                              

I’m having difficulty getting this code to work properly. The problem areas are:
1.) Recreating this line: newx,newy = mgrid[-1:1:100j,-1:1:100j]. In my case I have a dictionary with the grid in vector form. I’ve tried to recreate this line using np.meshgrid but then I get an error on line coords = scipy.array([ivals, jvals]). I’m looking for some help in recreating this Cookbook function and making it more dynamic
any help is greatly appreciated.

/M

Answer 1

You should have a look at the documentation for map_coordinates. I don’t see where the actual data you are trying to interpolate is in your code. What I mean is, presumably you have some data input which is a function of x and y; i.e. input = f(x,y) that you want to interpolate. In the first example you show, this is the array fvals. This should be your first argument to map_coordinates.

For example, if the data you are trying to inperpolate is input, which should be a 2-dimensional array of shape (len(x),len(y)), then the interpolated data would be:

interpolated_data = map_coordinates(input, coords)