Mailman 3 July 2009 - SciPy-Dev

10 min to Skypecon
by David Goldsmith 14 Jul '09

14 Jul '09

1 0

Re: [SciPy-dev] generic.base
by David Goldsmith 14 Jul '09

14 Jul '09

Nevermind, just found that info in http://docs.scipy.org/numpy/docs/numpy-docs/reference/arrays.scalars.rst/. Sorry for the noise. DG --- On Tue, 7/14/09, David Goldsmith <d_l_goldsmith(a)yahoo.com> wrote: > From: David Goldsmith <d_l_goldsmith(a)yahoo.com> > Subject: Re: [SciPy-dev] generic.base > To: "SciPy Developers List" <scipy-dev(a)scipy.org> > Date: Tuesday, July 14, 2009, 10:39 AM > > > --- On Tue, 7/14/09, Robert Kern <robert.kern(a)gmail.com> > > wrote: > > > > > The scalar types need to be immutable > > May I infer from this that all attributes of generic are > "read-only"? > > DG > > > > _______________________________________________ > Scipy-dev mailing list > Scipy-dev(a)scipy.org > http://mail.scipy.org/mailman/listinfo/scipy-dev >

1 0

Re: [SciPy-dev] generic.base
by David Goldsmith 14 Jul '09

14 Jul '09

> --- On Tue, 7/14/09, Robert Kern <robert.kern(a)gmail.com> > wrote: > > > The scalar types need to be immutable May I infer from this that all attributes of generic are "read-only"? DG

2 1

Re: [SciPy-dev] generic.base
by David Goldsmith 14 Jul '09

14 Jul '09

OK, thanks! DG --- On Tue, 7/14/09, Robert Kern <robert.kern(a)gmail.com> wrote: > From: Robert Kern <robert.kern(a)gmail.com> > Subject: Re: [SciPy-dev] generic.base > To: "SciPy Developers List" <scipy-dev(a)scipy.org> > Date: Tuesday, July 14, 2009, 12:06 AM > On Tue, Jul 14, 2009 at 00:50, David > Goldsmith<d_l_goldsmith(a)yahoo.com> > wrote: > > > > Hi! I'm working on the Scalar Base Class category > and, after a little bit of a shaky start, I think I have a > pretty good handle on it now. I wanted to see for my self > whether or not generic.base is "settable" (i.e., after > object creation), so I instantiated an np.complex64 object, > ala > > > > g = np.complex64(1) > > > > and then tried various ways (e.g., h=g.copy(), > h=g.view(), h=g.real, h=g.imag) to create a view of g (i.e., > create an object that shared its memory), but in all cases > h.base returned None, meaning, IIUC, h is _not_ a view of > (does not share memory with) g. So my questions are: are > objects derived from generic "viewable"; > > They are not. The scalar types need to be immutable and > allowing views > would make violating that too easy. > > > is so, how; if not, why the base property? > > So as to expose a consistent interface. > > -- > Robert Kern > > "I have come to believe that the whole world is an enigma, > a harmless > enigma that is made terrible by our own mad attempt to > interpret it as > though it had an underlying truth." > -- Umberto Eco > _______________________________________________ > Scipy-dev mailing list > Scipy-dev(a)scipy.org > http://mail.scipy.org/mailman/listinfo/scipy-dev >

1 0

generic.base
by David Goldsmith 14 Jul '09

14 Jul '09

Hi! I'm working on the Scalar Base Class category and, after a little bit of a shaky start, I think I have a pretty good handle on it now. I wanted to see for my self whether or not generic.base is "settable" (i.e., after object creation), so I instantiated an np.complex64 object, ala g = np.complex64(1) and then tried various ways (e.g., h=g.copy(), h=g.view(), h=g.real, h=g.imag) to create a view of g (i.e., create an object that shared its memory), but in all cases h.base returned None, meaning, IIUC, h is _not_ a view of (does not share memory with) g. So my questions are: are objects derived from generic "viewable"; is so, how; if not, why the base property? Thanks! DG

2 1

Re: [SciPy-dev] skype call agenda points
by d_l_goldsmith＠yahoo.com 14 Jul '09

14 Jul '09

Hi, Ralf, thanks for your thoughts - I hope you don't mind that I'm replying via the list: I didn't see anything in the email that looked like it should be kept private, and perhaps more people will participate if they see what and how much is on the agenda! :-) All your agenda items look good - not that I'd reject any of 'em, I see these things as open discussions, so all scipy-dev subscriber's concerns are fair game at these Skypecons. Also, thanks for RSVP-ing in the positive; others: please follow Ralf's example in that regard if you can. I have some comments on the items I'll give below, so here let me just remind everyone of the time: tomorrow, Tues. July 14, 18:00 UTC (an hour and a day earlier than usual). --- On Mon, 7/13/09, Ralf Gommers <ralf.gommers(a)googlemail.com> wrote: > Here are some thoughts on what to talk about during the > call tomorrow. I should be able to make it. > > - we're getting close to being done with Milestones > accessible to contributors without in-depth knowledge of > NumPy Awesome, thanks for pointing this out. > - how to get the specialized Milestones doc'ed (i.e. > ctypes, ufuncs, numpy internals,...) Well, I'll be acquiring "in-depth" knowledge as needed, but I certainly hope I don't have to do 'em all. > - what is next, scipy reference or numpy user guide, or > both? Open to discussion. > - SciPy Milestones page needs to be done Yup. > - note, scipy has 3000 docstrings, it took us >1 year to > do 1000 for numpy. how to ever get scipy done? Well, part of the reason it took so long is 'cause the Marathon stalled when the summer hired position terminated; I won't presume there's a causation there, but it certainly suggests that if Joe and I (mostly Joe) can secure long term funding, then the work can progress more continuously. > - review system, at least should show who reviewed what > docstring in the Log Agreed. Talk to you tomorrow, and again, thanks! DG PS: The other reminder - the "category" of the week is to make sure functions w/ nan equivalents are well-documented in that respect. > Cheers, > Ralf > > >

2 1

Finished off LA
by David Goldsmith 14 Jul '09

14 Jul '09

Linear Algebra status is now "Goal Met!" DG

1 0

Re: [SciPy-dev] Anyone wanna propose a category of the week and Skypecon reminder
by David Goldsmith 12 Jul '09

12 Jul '09

Darn, that means the queue is empty; thanks, Ralf. ;-) Oh wohl. OK, > > make sure that the NaN comparison and sorting issues are > > well-explained on all > > the functions that do comparison and sorting. Is now the "category" of the week. Thanks, Gael. Reminder: Skypecon this week is Tues. 18:00 UTC (instead of the usual Wed. 19:00 UTC). DG --- On Sat, 7/11/09, Ralf Gommers <ralf.gommers(a)googlemail.com> wrote: > From: Ralf Gommers <ralf.gommers(a)googlemail.com> > Subject: Re: [SciPy-dev] Anyone wanna propose a category of the week > To: "SciPy Developers List" <scipy-dev(a)scipy.org> > Date: Saturday, July 11, 2009, 10:49 PM > Actually, data type investigation is > done. I guess it is Sunday now so it counts:) > > Maybe we can do Gael's suggestion as well this week. > > Cheers, > Ralf > > > On Sun, Jul 12, 2009 at 12:20 AM, > David Goldsmith <d_l_goldsmith(a)yahoo.com> > wrote: > > > > Awesome, now we have one in the queue, that's what I > like to see! > > > > DG > > > > --- On Sat, 7/11/09, Gael Varoquaux <gael.varoquaux(a)normalesup.org> > wrote: > > > > > From: Gael Varoquaux <gael.varoquaux(a)normalesup.org> > > > Subject: Re: [SciPy-dev] Anyone wanna propose a > category of the week > > > To: "SciPy Developers List" <scipy-dev(a)scipy.org> > > > Date: Saturday, July 11, 2009, 8:48 PM > > > On Sat, Jul 11, 2009 > at 08:44:07PM > > > -0700, David Goldsmith wrote: > > > > > > > Data type investigation it is; we'll make the > week > > > Sunday to Saturday this week. Thanks, Ralf! > > > > > > Something that would also need some time (though not > as > > > much), is make > > > sure that the NaN comparison and sorting issues are > > > well-explained on all > > > the functions that do comparison and sorting. > > > > > > Gaël > > > _______________________________________________ > > > Scipy-dev mailing list > > > Scipy-dev(a)scipy.org > > > http://mail.scipy.org/mailman/listinfo/scipy-dev > > > > > > > > > > > _______________________________________________ > > Scipy-dev mailing list > > Scipy-dev(a)scipy.org > > http://mail.scipy.org/mailman/listinfo/scipy-dev > > > > > -----Inline Attachment Follows----- > > _______________________________________________ > Scipy-dev mailing list > Scipy-dev(a)scipy.org > http://mail.scipy.org/mailman/listinfo/scipy-dev >

1 0

Re: [SciPy-dev] Anyone wanna propose a category of the week
by David Goldsmith 12 Jul '09

12 Jul '09

You're too kind. ;-) DG --- On Sat, 7/11/09, Gael Varoquaux <gael.varoquaux(a)normalesup.org> wrote: > From: Gael Varoquaux <gael.varoquaux(a)normalesup.org> > Subject: Re: [SciPy-dev] Anyone wanna propose a category of the week > To: "SciPy Developers List" <scipy-dev(a)scipy.org> > Date: Saturday, July 11, 2009, 9:22 PM > On Sat, Jul 11, 2009 at 09:20:39PM > -0700, David Goldsmith wrote: > > > Awesome, now we have one in the queue, that's what I > like to see! > > It's great to have you around to manage the queue. > > Gaël > _______________________________________________ > Scipy-dev mailing list > Scipy-dev(a)scipy.org > http://mail.scipy.org/mailman/listinfo/scipy-dev >

1 0

Re: [SciPy-dev] Anyone wanna propose a category of the week
by David Goldsmith 12 Jul '09

12 Jul '09

Awesome, now we have one in the queue, that's what I like to see! DG --- On Sat, 7/11/09, Gael Varoquaux <gael.varoquaux(a)normalesup.org> wrote: > From: Gael Varoquaux <gael.varoquaux(a)normalesup.org> > Subject: Re: [SciPy-dev] Anyone wanna propose a category of the week > To: "SciPy Developers List" <scipy-dev(a)scipy.org> > Date: Saturday, July 11, 2009, 8:48 PM > On Sat, Jul 11, 2009 at 08:44:07PM > -0700, David Goldsmith wrote: > > > Data type investigation it is; we'll make the week > Sunday to Saturday this week. Thanks, Ralf! > > Something that would also need some time (though not as > much), is make > sure that the NaN comparison and sorting issues are > well-explained on all > the functions that do comparison and sorting. > > Gaël > _______________________________________________ > Scipy-dev mailing list > Scipy-dev(a)scipy.org > http://mail.scipy.org/mailman/listinfo/scipy-dev >

3 2