[Numpy-discussion] Maturing the Matrix class in NumPy

Stéfan van der Walt stefan@sun.ac...
Fri Jun 5 14:49:47 CDT 2009


Hi Alan

2009/6/5 Alan G Isaac <aisaac@american.edu>:
> You should take it very seriously when someone
> reports to you that the matrix object is a crucial to them,
> e.g., as a teaching tool.  Even if you do not find
> personally persuasive an example like
> http://mail.scipy.org/pipermail/numpy-discussion/2009-June/043001.html

You make a good point and the example you gave earlier shows how the
Matrix object can be used to write expressions that correspond clearly
to those in a textbook.

Christopher is right that most of the developers don't use the Matrix
class much, but that is no excuse to neglect it as we are currently
doing.  If the Matrix class is to remain, we need to take the steps
necessary to integrate it into NumPy properly.

To get going we'll need a list of changes required (i.e. "in an ideal
world, how would matrices work?").  There should be a set protocol for
all numpy functions that guarantees compatibility with ndarrays,
matrices and other derived classes.  This also ties in with a
suggestion by Darren Dale on ufuncs that hasn't been getting the
attention it deserves.  Indexing issues need to be cleared up, with a
chapter on matrix use written for the guide.

Being one of the most vocal proponents of the Matrix class, would you
be prepared to develop your Matrix Proposal at
http://scipy.org/NewMatrixSpec further?  We need to get to a point
where we can assign tasks, and start writing code.  The keyphrase here
should be "rough consensus" to avoid the bike shedding we've run into
in the past.

Thanks for following this through,

Regards
Stéfan


More information about the Numpy-discussion mailing list