[Numpy-discussion] remaining matrix-non-preserving functions

[Sven Schreiber]

Hi all,

I just discovered that the diff function returns a numpy-array even for
matrix inputs. Since I'm a card-carrying matrix fanatic, I hope that
behavior qualifies as a bug.

Then I went through some (most?) other functions/methods for which IMO
it's best to return matrices if the input is also a matrix-type. I found
that the following functions share the problem of diff (see below for
illustrations):

 vstack and hstack (although I always use r_ and c_ and they work fine
with matrices)

 outer

 msort


Should I open new tickets? (Or has this already been fixed since 0.9.8,
which I used because this time building the svn version failed for me?)

Cheers,
Sven

>>> n.__version__
'0.9.8'
>>> a
matrix([[1, 0, 0],
       [0, 1, 0],
       [0, 0, 1]])
>>> b
matrix([[0, 0, 0],
       [0, 0, 0]])
>>> n.diff(a)
array([[-1,  0],
       [ 1, -1],
       [ 0,  1]])
>>> n.outer(a,b)
array([[0, 0, 0, 0, 0, 0],
       [0, 0, 0, 0, 0, 0],
       [0, 0, 0, 0, 0, 0],
       [0, 0, 0, 0, 0, 0],
       [0, 0, 0, 0, 0, 0],
       [0, 0, 0, 0, 0, 0],
       [0, 0, 0, 0, 0, 0],
       [0, 0, 0, 0, 0, 0],
       [0, 0, 0, 0, 0, 0]])
>>> n.msort(a)
array([[0, 0, 0],
       [0, 0, 0],
       [1, 1, 1]])
>>> n.vstack([a,b])
array([[1, 0, 0],
       [0, 1, 0],
       [0, 0, 1],
       [0, 0, 0],
       [0, 0, 0]])
>>> n.hstack([a,b.T])
array([[1, 0, 0, 0, 0],
       [0, 1, 0, 0, 0],
       [0, 0, 1, 0, 0]])
>>>