On Sun, Nov 2, 2008 at 12:23 PM, Paul Rudin <paul@rudin.co.uk> wrote:

I'm experimenting with numpy and I've just written the code below, which
computes the thing I want (I think). Self.bits is an RxRxR array
representing a voxelized 3d model - values are either 0 or 1. I can't
help thinking that's there must be a much nicer way to do it. Any
suggestions?


 centre = numpy.array(scipy.ndimage.measurements.center_of_mass(self.bits))

 vectors = []
 for x in xrange(R):
   for y in xrange(R):
       for z in xrange(R):
           if self.bits[x,y,z]:
               vectors.append([x,y,z])

 vectors = numpy.array(vectors)
 distances = numpy.sqrt(numpy.sum((vectors-centre) ** 2.0, axis=1))
 av_dist = numpy.average(distances)

Try nonzero:

In [5]: bits = np.random.random_integers(0,1, size=(3,3,3))

In [6]: vectors = nonzero(bits)

In [7]: vectors
Out[7]:
(array([0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2]),
 array([0, 0, 0, 1, 1, 2, 2, 0, 0, 1, 2, 2, 0, 0, 1, 1, 2, 2]),
 array([0, 1, 2, 0, 1, 0, 2, 0, 2, 0, 1, 2, 0, 1, 0, 2, 0, 2]))

The arrays three arrays contain the x, y, z indices.

 Chuck