Mailman 3 February 2014 - NumPy-Discussion

ANN: Scipy 0.13.3 release
by Ralf Gommers Feb. 16, 2014

Feb. 16, 2014

Hi, I'm happy to announce the availability of the scipy 0.13.3 release. This is a bugfix only release; it contains fixes for regressions in ndimage and weave. Source tarballs can be found at https://sourceforge.net/projects/scipy/files/scipy/0.13.3/ and on PyPi. Release notes copied below, binaries will follow later (the regular build machine is not available for the next two weeks). Cheers, Ralf ========================== SciPy 0.13.3 Release Notes ========================== SciPy 0.13.3 is a bug-fix release with no new features compared to 0.13.2. Both the weave and the ndimage.label bugs were severe regressions in 0.13.0, hence this release. Issues fixed ------------ - 3148: fix a memory leak in ``ndimage.label``. - 3216: fix weave issue with too long file names for MSVC. Other changes ------------- - Update Sphinx theme used for html docs so ``>>>`` in examples can be toggled. Checksums ========= 0547c1f8e8afad4009cc9b5ef17a2d4d release/installers/scipy-0.13.3.tar.gz 20ff3a867cc5925ef1d654aed2ff7e88 release/installers/scipy-0.13.3.zip

1 1

Removal of doc/cython and doc/pyrex
by Charles R Harris Feb. 16, 2014

Feb. 16, 2014

Apropos #2384 <https://github.com/numpy/numpy/issues/2384>, Ralph has suggested removing the doc/pyrex directory as irrelevant at this point. Along the same lines, we could remove doc/cython as cython now has good numpy support. There is also doc/swig, which contains numpy.i, and which we probably need to retain, but maybe put somewhere else. Thoughts? Chuck

1 0

Requesting Code Review of nanmedian ENH
by David Freese Feb. 16, 2014

Feb. 16, 2014

Hi everyone, I put together a np.nanmedian function to extend np.median to handle nans. Could someone review this code and give me some feedback on it before I submit a pull request for it? https://github.com/dfreese/numpy/compare/master...feature;nanmedian Thanks, David

3 5

ANN: SciPy 2014 Conference, July 6th - 12th, Austin, TX!
by Anthony Scopatz Feb. 16, 2014

Feb. 16, 2014

Hello All! I am pleased to announce that *SciPy 2014*, the thirteenth annual *Scientific Computing with Python conference*, will be held this July 6th-12th in Austin, Texas. SciPy is a community dedicated to the advancement of scientific computing through open source Python software for mathematics, science, and engineering. The annual SciPy Conference allows participants from all types of organizations to showcase their latest projects, learn from skilled users and developers, and collaborate on code development. For more information please visit our website: https://conference.scipy.org/scipy2014/ This year the conference has been extended to include an additional day of presentations. During the presentation days SciPy is proud to host the following event and talk types: - Keynotes - Expert Panels - Short Talks - Poster Presentations - Birds of a Feather Sessions The full program will consist of two days of tutorials by followed by three days of presentations, and concludes with two days of developer sprints on projects of interest to attendees. This year, we are excited to present a job fair for the first time! <https://github.com/scipy-conference/scipy-conference/blob/master/mailings/c…>Specialized Tracks This year we are happy to announce two specialized tracks that run in parallel to the general conference: *Scientific Computing in Education* Thanks to efforts such as Software Carpentry, the Hacker Within and grassroots Python Bootcamps, teaching scientific computing as a discipline is becoming more widely accepted and recognized as a crucial task in developing scientific literacy. This special track will focus on efforts to promote and develop scientific computing education, as well as related topics such as reproducibility and best practices for scientific computing. *Geospatial Data in Science* Python has become a core component of organiziing, understanding, and visualizing geospatial data. This track will focus on libraries, tools and techniques for processing Geospatial data of all types and for all purposes -- from low-volume to high-volume, local and global. <https://github.com/scipy-conference/scipy-conference/blob/master/mailings/c…>Domain-specific Mini-symposia Introduced in 2012, mini-symposia are held to discuss scientific computing applied to a specific scientific domain/industry during a half afternoon after the general conference. Their goal is to promote industry specific libraries and tools, and gather people with similar interests for discussions. Mini-symposia on the following topics will take place this year: - Astronomy and astrophysics - Bioinformatics - Geophysics - Vision, Visualization, and Imaging - Computational Social Science and Digital Humanities - Engineering <https://github.com/scipy-conference/scipy-conference/blob/master/mailings/c…> Tutorials Multiple interactive half-day tutorials will be taught by community experts. The tutorials provide conceptual and practical coverage of tools that have broad interest at both an introductory or advanced level. This year, a third track will be added, which target specifically programmers with no prior knowledge of scientific python. <https://github.com/scipy-conference/scipy-conference/blob/master/mailings/c…>Developer Sprints A hackathon environment is setup for attendees to work on the core SciPy packages or their own personal projects. The conference is an opportunity for developers that are usually physically separated to come together and engage in highly productive sessions. It is also an occasion for new community members to introduce themselves and receive tips from community experts. This year, some of the sprints will be scheduled and announced ahead of the conference. <https://github.com/scipy-conference/scipy-conference/blob/master/mailings/c…>Birds-of-a-Feather (BOF) Sessions Birds-of-a-Feather sessions are self-organized discussions that run parallel to the main conference. The BOFs sessions cover primary, tangential, or unrelated topics in an interactive, discussion setting. This year, some of the BOF sessions will be scheduled and announced ahead of the conference. <https://github.com/scipy-conference/scipy-conference/blob/master/mailings/c…>Important Dates - March 14th: Presentation abstracts, poster, tutorial submission deadline. Application for sponsorship deadline. - April 17th: Speakers selected - April 22nd: Sponsorship acceptance deadline - May 1st: Speaker schedule announced - May 6th, or 150 registrants: Early-bird registration ends - July 6-12th: 2 days of tutorials, 3 days of conference, 2 days of sprints We look forward to a very exciting conference and hope to see you all in Austin this summer! The SciPy2014 Organizers

1 0

bool value of dtype is False?
by Charles G. Waldman Feb. 15, 2014

Feb. 15, 2014

>>> d = numpy.dtype(int) >>> if d: print "OK" ... else: print "I'm surprised" I'm surprised

2 1

deprecate numpy.matrix
by alex Feb. 13, 2014

Feb. 13, 2014

Hello list, I wrote this mini-nep for numpy but I've been advised it is more appropriate for discussion on the list. """ The ``numpy.matrix`` API provides a low barrier to using Python for linear algebra, just as the pre-3 Python ``input`` function and ``print`` statement provided low barriers to using Python for automatically evaluating input and for printing output. On the other hand, it really needs to be deprecated. Let's deprecate ``numpy.matrix``. """ I understand that numpy.matrix will not be deprecated any time soon, but I hope this will register as a vote to help nudge its deprecation closer to the realm of acceptable discussion. Alex

17 74

FYI: libflatarray
by Neal Becker Feb. 13, 2014

Feb. 13, 2014

I thought this was interesting: http://www.libgeodecomp.org/libflatarray.html

3 2

Geometrically defined masking arrays; how to optimize?
by Wolfgang Draxinger Feb. 12, 2014

Feb. 12, 2014

Hi, I implemented the following helper function to create masking arrays: def gen_mask(self, ring, sector): in_segment = None if ring == 0: radius = self.scales()[0] def in_center_circle(xy): dx = xy[0] - self.xy[0] dy = xy[1] - self.xy[1] r = math.sqrt( dx**2 + dy**2 ) return r < radius in_segment = in_center_circle else: angle = ( self.a_sector(sector, ring), self.a_sector( (sector+1) % self.n_sectors(), ring) ) radii = self.scales()[ring:ring+1] def in_segment_ring(xy): dx = xy[0] - self.xy[0] dy = xy[1] - self.xy[1] r = math.sqrt( dx**2 + dy**2 ) a = math.atan2( dy, dx ) return r >= radii[0] and r < radii[1] \ and a >= angle[0] and a < angle[1] in_segment = in_segment_ring width,height = self.arr.shape mask = numpy.zeros(shape=(width, height), dtype=numpy.bool) for y in range(height): for x in range(width): mask[x][y] = in_segment((x,y)) return mask self.scales() generates a list of radius scaling factors. self.a_sector() gives the dividing angle between sectors "sector" and "sector+1" on the given ring. The function works, it generates the masks as I need it. The problem is - of course - that it's quite slow due to the inner loops the perform the geometrical test if the given element of the array self.arr is withing the bounds of the ring-sector for which the mask is generated. I wonder if you guys have some ideas, on how I could accelerate this. This can be perfectly well constructed by boolean combination of elementary geometrical objects. For example a ring would be ring(p, r0, r1): disk(p, r1) xor disk(p, r0) # where r0 < r1 The sector would then be ring_sector(p, r, s): ring(p, r, r + ...) and sector(p, s, s+1) I'd like to avoid doing this using some C helper routine. I'm looking for something that is the most efficient when it comes to "speedup"/"time invested to develop this". Cheers, Wolfgang

4 5

Overlapping time series
by Daniele Nicolodi Feb. 12, 2014

Feb. 12, 2014

Hello, I have two time series (2xN dimensional arrays) recorded on the same time basis, but each with it's own dead times (and start and end recording times). I would like to obtain two time series containing only the time overlapping segments of the data. Does numpy or scipy offer something that may help in this? I can imagine strategies about how to approach the problem, but none that would be efficient. Ideas? Thank you. Best, Daniele

5 16

record arrays with char*?
by Christopher Jordan-Squire Feb. 11, 2014

Feb. 11, 2014

I'm trying to wrap some C code using cython. The C code can take inputs in two modes: dense inputs and sparse inputs. For dense inputs the array indexing is naive. I have wrappers for that. In the sparse case the matrix entries are typically indexed via names. So, for example, the library documentation includes this as input you could give: struct { char* ind; double val, wght; } data[] = { {"camera", 15, 2}, {"necklace", 100, 20}, {"vase", 90, 20}, {"pictures", 60, 30}, {"tv", 40, 40}, {"video", 15, 30}}; At the C level, data is passed to the function by directly giving its address. (i.e. the C function takes as an argument (unsigned long) data, casting the data pointer to an int) I'd like to create something similar using record arrays, such as np.array([("camera", 15, 2), ("necklace", 100, 20), ... ], dtype='object,f8,f8'). Unfortunately this fails because (1) In cython I need to determine the address of the first element and I can't take the address of a an input whose type I don't know (the exact type will vary on the application, so more or fewer fields may be in the C struct) (2) I don't think a python object type is what I want--I need a char* representation of the string. (Unfortunately I can't test this because I haven't solved (1) -- how do you pass a record array around in cython and/or take its address?) Any suggestions?

2 3