[SciPy-user] Direct sum in numpy/scipy ?

[Bill Baxter]

On 5/19/06, Nils Wagner <nwagner at iam.uni-stuttgart.de> wrote:Hi all,

> How can I build a direct sum of two matrices in numpy/scipy ?

Is there something wrong with just allocating the target matrix and
assigning the parts to it?
Like so:
  dsum = num.zeros( num.add(a.shape,b.shape) )
  dsum[:a.shape[0],:a.shape[1]]=a
  dsum[a.shape[0]:,a.shape[1]:]=b


Full example:
>>> import numpy as num
>>> a=num.arange(0,12).reshape(3,4)
>>> b=num.arange(0,6).reshape(3,2)
>>> a
array([[ 0,  1,  2,  3],
       [ 4,  5,  6,  7],
       [ 8,  9, 10, 11]])
>>> b
array([[0, 1],
       [2, 3],
       [4, 5]])


>>> dsum = num.zeros( num.add(a.shape,b.shape) )
>>> dsum.shape
(6, 6)
>>> dsum[:a.shape[0],:a.shape[1]]=a
>>> dsum[a.shape[0]:,a.shape[1]:]=b
>>> dsum
array([[ 0,  1,  2,  3,  0,  0],
       [ 4,  5,  6,  7,  0,  0],
       [ 8,  9, 10, 11,  0,  0],
       [ 0,  0,  0,  0,  0,  1],
       [ 0,  0,  0,  0,  2,  3],
       [ 0,  0,  0,  0,  4,  5]])


Or maybe you wanted a sparse result?

--bb
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.scipy.org/pipermail/scipy-user/attachments/20060519/9ac0c7cc/attachment.html>