[SciPy-User] scipy.sparse.csr_matrix.matmat deprecation question

Nathan Bell wnbell@gmail....
Sun Jun 27 14:05:34 CDT 2010


On Fri, Jun 25, 2010 at 1:09 PM, Andrew Schein <andrew@andrewschein.com> wrote:
> I would like to perform a matrix multiplication of the form
>
> A * B
>
> where A is dense and B is sparse CSR or COO.  Does scipy.sparse have this
> capability and will it in the future?  How fast is the scipy implementation
> in comparison to INTEL MKL?
>
> It appears that there is a .matmat function that has been deprecated.  Does
> this reflect a retreat, or is the functionality found in some other place?
>
> Thanks,
>

Hi Andrew,

All sparse matrix multiplication functionality is exposed via
__mul__() now, so the matmat function is unnecessary.  Simply using
A*B should do the appropriate thing.

I don't know how the speed compares to MKL, but the code is
implemented in C++ so it should be reasonably fast.

-- 
Nathan Bell wnbell@gmail.com
http://www.wnbell.com/


More information about the SciPy-User mailing list