[Numpy-discussion] 2D binning

[Wes McKinney]

On Wed, Jun 2, 2010 at 6:18 PM, Stephen Simmons <mail at stevesimmons.com> wrote:
>
> On 1/06/2010 10:51 PM, Wes McKinney wrote:
>  > <snip>
>  > This is a pretty good example of the "group-by" problem that will
>  > hopefully work its way into a future edition of NumPy.
>
> Wes (or anyone else), please can you elaborate on any plans for groupby?
>
> I've made my own modification to numpy.bincount for doing groupby-type
> operations but not contributed anything back to the numpy community.
>
> Is there any interest from European-based people to work on groupby etc
> at the Euro SciPy sprints in July? If others are interested, maybe we
> could work out requirements beforehand and then do some coding in Paris.
>
> Cheers
> Stephen
>
> _______________________________________________
> NumPy-Discussion mailing list
> NumPy-Discussion at scipy.org
> http://mail.scipy.org/mailman/listinfo/numpy-discussion
>

I know we're planning to discuss at SciPy in Austin and will hopefully
have a gameplan. We should coordinate the discussions / implementation
somewhere. I've implemented some groupby functionality in pandas but I
guess at a higher level than what's been proposed to add to NumPy.

On the pystatsmodels mailing list we've been starting to have some
discussions about statistically-oriented data structures in general--
I'll be sending an e-mail to the NumPy mailing list soon to start a
broader discussion which will hopefully lead to some good things at
the conferences.