[Numpy-discussion] Scipy 2017 NumPy sprint

Sebastian Berg sebastian at sipsolutions.net
Sun Jul 2 11:33:19 EDT 2017


On Sun, 2017-07-02 at 10:49 -0400, Allan Haldane wrote:
> On 07/02/2017 10:03 AM, Charles R Harris wrote:
> > Updated list below.
> > 
> > On Sat, Jul 1, 2017 at 7:08 PM, Benjamin Root <ben.v.root at gmail.com
> >  
> > <mailto:ben.v.root at gmail.com>> wrote:
> > 
> >     Just a heads-up. There is now a sphinx-gallery plugin.
> > Matplotlib
> >     and a few other projects have migrated their docs over to use
> > it.
> > 
> >     https://sphinx-gallery.readthedocs.io/en/latest/
> >     <https://sphinx-gallery.readthedocs.io/en/latest/>
> > 
> >     Cheers!
> >     Ben Root
> > 
> > 
> >     On Sat, Jul 1, 2017 at 7:12 AM, Ralf Gommers <ralf.gommers at gmai
> > l.com
> >     <mailto:ralf.gommers at gmail.com>> wrote:
> > 
> > 
> > 
> >         On Fri, Jun 30, 2017 at 6:50 AM, Pauli Virtanen <pav at iki.fi
> >         <mailto:pav at iki.fi>> wrote:
> > 
> >             Charles R Harris kirjoitti 29.06.2017 klo 20:45:
> >             >     Here's a random idea: how about building a NumPy
> > gallery?
> >             >     scikit-{image,learn} has it, and while those
> > projects may have more
> >             >     visual datasets, I can imagine something along
> > the lines of Nicolas
> >             >     Rougier's beautiful book:
> >             >
> >             >     http://www.labri.fr/perso/nrougier/from-python-to
> > -numpy/
> >             <http://www.labri.fr/perso/nrougier/from-python-to-nump
> > y/>
> >             >     <http://www.labri.fr/perso/nrougier/from-python-t
> > o-numpy/
> >             <http://www.labri.fr/perso/nrougier/from-python-to-nump
> > y/>>
> >             >
> >             >
> >             > So that would be added in the  numpy
> >             > <https://github.com/numpy>/numpy.org
> > <http://numpy.org>
> >             > <https://github.com/numpy/numpy.org
> >             <https://github.com/numpy/numpy.org>> repo?
> > 
> >             Or https://scipy-cookbook.readthedocs.io/
> >             <https://scipy-cookbook.readthedocs.io/>  ?
> >             (maybe minus bitrot and images added :)
> >             _____________________________________
> > 
> > 
> >         I'd like the numpy.org <http://numpy.org> one. numpy.org
> >         <http://numpy.org> is now incredibly sparse and ugly, a
> > gallery
> >         would make it look a lot better.
> > 
> >         Another idea, from the "deprecate np.matrix" discussion:
> > add
> >         numpy documentation describing the preferred way to handle
> >         matrices, extolling the virtues of @, and move np.matrix
> >         documentation to a deprecated section.
> > 
> > 
> >   Putting things together with a few new ideas,
> > 
> >  1. add gallery to numpy.org <http://numpy.org>,
> >  2. add extended documentation of '@' operator,
> >  3. make Numpy tests Pytest compatible,
> >  4. add matrix multiplication ufunc.
> > 
> >   Any more ideas?
> 
> The new doctest runner suggested in the printing thread? This is to 
> ignore whitespace and precision in ndarray output.
> 
> I can see an argument for distributing it in numpy if it is designed
> to 
> be specially aware of ndarrays or numpy scalars (eg to test equality 
> between 'wants' and 'got')
> 

I don't really feel it is very numpy specific or should be under the
numpy umbrella (I mean if there is no other spot, I guess it could live
on the numpy github page). Its about as numpy specific, as the gallery
sphinx extension is probably matplotlib specific....

That doesn't mean that it might not be a good sprint, though :).

The question to me is a bit what those who actually go there want from
it or do a few people who know numpy/scipy already plan to come? Two
years ago, we did not have much of a plan, so it was mostly giving
three people or so a bit of a tutorial of how numpy worked internally
leading to some bug fixes.

One quick idea that might be nice and dives a bit into the C-layer
(might be nice if there is no big topic with a few people working on):

* Find places that should have the new memory overlap
  detection and implement it there.

If someone who does subclasses/array-likes or so (e.g. like Stefan
Hoyer ;)) and is interested, and also we do some
teleconferencing/chatting (and I have time).... I might be interested
in discussing and possibly trying to develop the new indexer ideas,
which I feel are pretty far, but I got stuck on how to get subclasses
right.

- Sebastian


> Allan
> 
> 
> > Chuck
> > 
> > 
> > 
> > _______________________________________________
> > NumPy-Discussion mailing list
> > NumPy-Discussion at python.org
> > https://mail.python.org/mailman/listinfo/numpy-discussion
> > 
> 
> _______________________________________________
> NumPy-Discussion mailing list
> NumPy-Discussion at python.org
> https://mail.python.org/mailman/listinfo/numpy-discussion
> 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 801 bytes
Desc: This is a digitally signed message part
URL: <http://mail.python.org/pipermail/numpy-discussion/attachments/20170702/9f547777/attachment.sig>


More information about the NumPy-Discussion mailing list