Numerical Python
	================

Numerical Python adds a fast, compact, multidimensional array language 
facility to Python.  

 o Complete information is available at http://www.pfdubois.com/numpy/

 o Data types for arrays include byte, integer, float, double and complex.

 o Arrays support a variety of python style slicing operations to access
   subregions of the array. 

 o Numpy includes interfaces to the LAPACK linear algebra library and the
   FFTPACK Fourier transforms library. 

 o Convenient methods have been adding for translating GDAL images into 
   NumPy arrays, and for displaying NumPy arrays in OpenEV.  However, the
   NumPy representation loses georeferencing, metadata, and colortables. 

This example loads a band from a GDAL supported file (dist_file) into a
NumPy array (distmap), converts it to floating point, does some manipulations,
displays the result and then saves to another file. 

   from Numeric import *
   from gdalnumeric import *

   distmap = LoadFile( dist_file )
   # hack: make sure no values are zero
   distmap = distmap.astype(Float32) + 0.001
   weightmap = sample[3] / power (distmap , exponent)
   weightmap = weightmap.astype(Float32)
   display( weightmap )
   SaveArray( weightmap, weight_file, 'PAux' )

(demo similar steps in OpenEV using the Interactive Shell)   

It would be good to have a more in depth review of NumPy another time.