[Numpy-discussion] Proposal for matrix_rank function in numpy

[Matthew Brett]

Hi,

On Wed, Dec 16, 2009 at 2:16 PM, Gael Varoquaux
<gael.varoquaux at normalesup.org> wrote:
> On Wed, Dec 16, 2009 at 02:13:08PM -0500, Matthew Brett wrote:
>> I'm happy to write the doctests as tests.   My feeling is there is no
>> objection to this function at the moment, so it would be reasonable,
>> unless I hear otherwise, to commit to SVN.
>
> I have one small comment: I am really happy to see you working on this
> function. It will be very useful, and its great to have it in a
> generaly-available package. However, I wonder whether it belongs to
> numpy.linalg, or scipy.linalg. I was under the impression that we should
> direct users who have linalg problems to scipy, as it can do much more.

It's another option, and one I thought of, but in this case, the use
is so ubiquitous in linear algebra, and the function so small, that it
would seem a pity to require installing scipy to get it.

See you,

Matthew