Re: [Numpy-discussion] Universal functions and introspection

I don't think the names exist currently except as mnemonics in the docstring. Patches to improve 'inspect' support would be welcome of course. -n On Tue, Apr 15, 2014 at 11:30 PM, Sylvain Corlay wrote: > Hi, > Unfortunately, numpy ufuncs are incompatible with the inspect module. > While ufunc.ni

[Numpy-discussion] Generalized universal functions returning higher-dimensional arrays.

Hello all, numpy.vectorize is a very convenient way to build universal function from scalar functions, and handle broadcasting... A use case that is not handled is the generalization to functions already returning nd array: For example: If foo(x,y) is a scalar function returning a array of dim k,

[Numpy-discussion] Universal functions and introspection

Hi, Unfortunately, numpy ufuncs are incompatible with the inspect module. While ufunc.nin and ufunc.nout provide some information on the signature, I don't see how to retrieve argument names. What is the preferred/best way to get argument names of a numpy ufuncs? Best, Sylvain _

[Numpy-discussion] EuroSciPy 2014 abstract submission deadline extended

Hi, In response to a number of requests, the organizing committee of the EuroSciPy 2014 conference has extended the deadline for abstract submission by 12 days, to Sunday April 27th 2014, 23:59:59 UTC. Up to then, new abstracts may be submitted on https://www.euroscipy.org/2014/ . We are very much

Re: [Numpy-discussion] High-quality memory profiling for numpy in python 3.5 / volunteers needed

On 15.04.2014 18:39, Nathaniel Smith wrote: > On Tue, Apr 15, 2014 at 4:08 PM, Julian Taylor > mailto:jtaylor.deb...@googlemail.com>> > wrote: >> On Tue, Apr 15, 2014 at 3:07 PM, Nathaniel Smith > wrote: >>> On Tue, Apr 15, 2014 at 12:06 PM, Julian Taylor >>>

Re: [Numpy-discussion] High-quality memory profiling for numpy in python 3.5 / volunteers needed

On 15.04.2014 18:39, Nathaniel Smith wrote: > On Tue, Apr 15, 2014 at 4:08 PM, Julian Taylor > mailto:jtaylor.deb...@googlemail.com>> > wrote: >> On Tue, Apr 15, 2014 at 3:07 PM, Nathaniel Smith > wrote: >>> On Tue, Apr 15, 2014 at 12:06 PM, Julian Taylor >>>

Re: [Numpy-discussion] Wiki page for building numerical stuff onWindows

Hi, On Tue, Apr 15, 2014 at 11:14 AM, Aron Ahmadia wrote: > Fernando is referring to this: http://lightningpython.org/ > > I have to admit that as a Pythonist supporting Windows builds, Cygwin and/or > MinGW have always been the most appealing because so much of the rest of the > numerical comput

Re: [Numpy-discussion] Wiki page for building numerical stuff onWindows

Hi, On Tue, Apr 15, 2014 at 11:10 AM, Fernando Perez wrote: > > Quick question on this: has anyone looked at the Open Watcom compiler? There > was a lighting talk at Pycon about a guy who's been working on getting > Python itself to build on Windows with this compiler. I don't know if it > might

Re: [Numpy-discussion] Wiki page for building numerical stuff onWindows

Fernando is referring to this: http://lightningpython.org/ I have to admit that as a Pythonist supporting Windows builds, Cygwin and/or MinGW have always been the most appealing because so much of the rest of the numerical computing infrastructure is primarily supported on UNIXy platforms. A On

Re: [Numpy-discussion] Wiki page for building numerical stuff onWindows

Quick question on this: has anyone looked at the Open Watcom compiler? There was a lighting talk at Pycon about a guy who's been working on getting Python itself to build on Windows with this compiler. I don't know if it might help in all this, if nothing else it mitght be good to have. This was th

Re: [Numpy-discussion] High-quality memory profiling for numpy in python 3.5 / volunteers needed

On Tue, Apr 15, 2014 at 4:08 PM, Julian Taylor < jtaylor.deb...@googlemail.com> wrote: > On Tue, Apr 15, 2014 at 3:07 PM, Nathaniel Smith wrote: >> On Tue, Apr 15, 2014 at 12:06 PM, Julian Taylor >> wrote: Good news, though! python-dev is in favor of adding calloc() to the core allocati

Re: [Numpy-discussion] High-quality memory profiling for numpy in python 3.5 / volunteers needed

On Tue, Apr 15, 2014 at 3:07 PM, Nathaniel Smith wrote: > On Tue, Apr 15, 2014 at 12:06 PM, Julian Taylor > wrote: >>> Good news, though! python-dev is in favor of adding calloc() to the >>> core allocation interfaces, which will let numpy join the party. See >>> python-dev thread: >>> https://

Re: [Numpy-discussion] High-quality memory profiling for numpy in python 3.5 / volunteers needed

On Tue, Apr 15, 2014 at 12:06 PM, Julian Taylor wrote: >> Good news, though! python-dev is in favor of adding calloc() to the >> core allocation interfaces, which will let numpy join the party. See >> python-dev thread: >> https://mail.python.org/pipermail/python-dev/2014-April/133985.html >> >>

Re: [Numpy-discussion] High-quality memory profiling for numpy in python 3.5 / volunteers needed

> Good news, though! python-dev is in favor of adding calloc() to the > core allocation interfaces, which will let numpy join the party. See > python-dev thread: > https://mail.python.org/pipermail/python-dev/2014-April/133985.html > > It would be especially nice if we could get this into 3.5, si

[Numpy-discussion] High-quality memory profiling for numpy in python 3.5 / volunteers needed

Hey all, The well known memory_profiler module [1] is super-useful, but has a fundamental limitation, which is the only way it can track allocations is by constantly polling the OS for the size of the total process address space. This is a crude and unreliable way of making measurements. In Pytho

Re: [Numpy-discussion] index partition

On 14 April 2014 18:17, Alan G Isaac wrote: > I find it rather more convenient to use boolean arrays, > but I wonder if arrays of indexes might have other > advantages (which would suggest using the set operations > instead). In particular, might a[boolean_array] be slower > that a[indexes]? (I'

Re: [Numpy-discussion] Test error with ATLAS, Windows 64 bit

On Tue, Apr 15, 2014 at 4:30 AM, Matthew Brett wrote: > > It looks as though mingw-w64 is at fault, and I was confused (still > am) because of the different behavior with double and a constant: > > #include > #include > > int main() { > double z, i = -0.0; > printf("With double %f=%f, wi