On May 19, 12:08 pm, William Stein <[email protected]> wrote:
> On Wed, May 19, 2010 at 8:54 AM, kcrisman <[email protected]> wrote:
>
> > On May 19, 11:36 am, Jason Grout <[email protected]> wrote:
> >> On 05/19/2010 10:22 AM, William Stein wrote:
>
> >> > On Wednesday, May 19, 2010, Jason Grout<[email protected]>  
> >> > wrote:
> >> >> On 05/19/2010 09:58 AM, Tobias Katz wrote:
> >> >> Is the best way to convolve numerically?
>
> >> >> You might look at scipy or numpy (both of which are included in Sage). 
> >> >> They have functions which do convolutions.
>
> >> > I think the OP asked for *symbolic* convolution.   I don't think
> >> > scipy/jumpy do anything symbolically.
>
> >> I agree.  I was responding to his last question about convolving things
> >> numerically.
>
> > Symbolic is implemented in Mma 
> > (seehttp://mathworld.wolfram.com/Convolution.html)
> > but Maple seems to only have it for audio applications (? based on
> > brief search on maplesoft.com).  Should we open a ticket for this?
>
> Yes, based on the mission statement of Sage: "... viable alternative
> to ... Mathematica"

Sorry, I meant "is there already something related open, or is this
already implemented"...

And it turns out that there is a whole file (by David Harvey)
sage.rings.polynomial.convolution, as well as the method
sage.functions.piecewise.PiecewisePolynomial.convolution, not to
mention sage.gsl.dft.IndexedSequence.convolution (both of which are
due to David Joyner, who has already replied on this thread!).

So what do we *not* have implemented?  Looks like polynomial
convolution is good, if hard to find, and also discrete/Dirichlet
convolution for IndexedSequences.  How hard would it be to extend
these things - or would we want a totally new implementation for SR,
and for arithmetic functions?

This is now http://trac.sagemath.org/sage_trac/ticket/8994 .  I also
hope that Tobias will find at least one of the already existing
functions useful, perhaps the piecewise implementation if he has
compact support?

- kcrisman

-- 
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to 
[email protected]
For more options, visit this group at 
http://groups.google.com/group/sage-support
URL: http://www.sagemath.org

Reply via email to