Mailman 3 September 2006 - SciPy-Dev

Question about Kaiser Implementation in firwin
by Buehler, Eric (AGRE) Oct. 4, 2008

Oct. 4, 2008

Hello, I have been using the signal.firwin FIR design tool and I have a question about the implementation. Scipy-0.4.8 Numpy-0.9.6 I am generating a kaiser window with the following parameters: N = 128 Cutoff = ~4e7 width = .3 All of the special kaiser generation parameters in the function follow Oppenheim and Schafer well. Even the sinc function for linear phase is follows exactly. However, the last line of the firwin function is causing me some heartburn. filter_design.py 1538 win = get_window(window,N,fftbins=1) 1539 alpha = N//2 1540 m = numpy.arange(0,N) 1541 h = win*special.sinc(cutoff*(m-alpha)) 1542 return h / sum(h) Line 1542 of filter_design.py, "return h / sum(h)", normalizes the function where it doesn't seem necessary, at least in the kaiser window case. Without the normalization, the kaiser window already returns a value of 1 at the zero frequency point. This normalization scales all of the data, making the window difficult to use in the frequency domain. Can someone point me to the rationale for this line? Looking at the code, this seems to be a pretty recent change (within the last year/year and a half). Thanks, Eric Buehler eric <dot> buehler <at> smiths-aerospace <dot> com ****************************************** The information contained in, or attached to, this e-mail, may contain confidential information and is intended solely for the use of the individual or entity to whom they are addressed and may be subject to legal privilege. If you have received this e-mail in error you should notify the sender immediately by reply e-mail, delete the message from your system and notify your system manager. Please do not copy it for any purpose, or disclose its contents to any other person. The views or opinions presented in this e-mail are solely those of the author and do not necessarily represent those of the company. The recipient should check this e-mail and any attachments for the presence of viruses. The company accepts no liability for any damage caused, directly or indirectly, by any virus transmitted in this email. ******************************************

3 2

umfpack-5.1
by Darren Dale Oct. 6, 2006

Oct. 6, 2006

After updating umfpack to the current version (umfpack-5.1), I am not able to build scipy. There is a long list of errors, such as /usr/include/umfpack/umfpack.h:31:22: error: UFconfig.h: No such file or directory In file included from /usr/include/umfpack/umfpack.h:48, from build/src.linux-x86_64-2.4/Lib/linsolve/umfpack/_umfpack_wrap.c:2478: /usr/include/umfpack/umfpack_symbolic.h:23: error: expected '=', ',', ';', 'asm' or '__attribute__' before 'umfpack_dl_symbolic' /usr/include/umfpack/umfpack_symbolic.h:47: error: expected '=', ',', ';', 'asm' or '__attribute__' before 'umfpack_zl_symbolic' and build/src.linux-x86_64-2.4/Lib/linsolve/umfpack/_umfpack_wrap.c:2810: error: 'UF_long' undeclared (first use in this function) build/src.linux-x86_64-2.4/Lib/linsolve/umfpack/_umfpack_wrap.c:2810: error: (Each undeclared identifier is reported only once build/src.linux-x86_64-2.4/Lib/linsolve/umfpack/_umfpack_wrap.c:2810: error: for each function it appears in.) build/src.linux-x86_64-2.4/Lib/linsolve/umfpack/_umfpack_wrap.c:2810: error: expected ';' before 'arg1' Will scipy support new versions of umfpack? Thanks, Darren

2 2

Online SciPy Journal
by Travis Oliphant Oct. 4, 2006

Oct. 4, 2006

As most of you know, I'm at an academic institution and have a strong belief that open source should mesh better with the academic world then it currently does. One of the problems, is that it is not easy to just point to open source software as a "publication" which is one of the areas that most tenure-granting boards look at when deciding on a candidate. To help with that a little-bit, I'd really like to get a peer-reviewed on-line journal for code contributed to SciPy started. Not to implicate him if he no longer thinks it's a good idea, but I first started thinking seriously about this at SciPy 2006 when Fernando Perez mentioned he had seen this model with a math-related software package and thought we could try something similar. Exactly what form that will take is worth some discussion. Most "publications" are viewed highly because they are "peer-reviewed". It seems to me, that we have all the makings of a community that could peer-review code contributed to SciPy. I'd like to see two parts to "the journal" (maybe they are separate). One is the code itself. Not all of the code currently in SciPy is "journal" quality (I look no further than many of my own contributions...) We would need a way to specify what is "published" SciPy and what is not (besides the sandbox mechansim). The other aspect to the SciPy Journal is an "article" that goes into detail about some particular code contributed to SciPy. Not all code will need this, but many contributions will. This is where the design decisions are discussed and recorded. Such a project needs many associate editors who would help find reviewers and make decisions about publication. If this project works, it is one means that would allow me to spend more time on SciPy and "have it count" toward academic rank and status. I have not thought it through well enough to have the details ironed out, but I'm soliciting feedback from you who would likely be involved. Besides your comments, I would like a response on at least two questions: +1, -1, +0, -0 voting. 1) Do you like the idea ? 2) Could you help (i.e. as an editor, reviewer, etc.)?

16 34

Intro / Docstrings
by Greg Willden Sept. 30, 2006

Sept. 30, 2006

Hello all, The FAQ says if you want to help out drop you line. Here is my line: I'm pretty new to numpy/scipy but I've been using Python for a while now. I have a long history of Matlab use but I'm gradually weaning myself. I would like to contribute to this project by adding a few things to the docstrings that will help new people like myself get their bearings. I love Python and would really like to get really good with numpy/scipy/matplotlib. I want to see more "See also:" portions on the end of the docstrings. This is something that Matlab does well that makes it easy to learn about new, related functions. What is the best way for me to contribute? Whom would I submit patches to? This list or an individual? Thanks Greg -- Linux. Because rebooting is for adding hardware.

3 3

from scipy import maxentropy fails with Python 2.5
by Lars Bittrich Sept. 30, 2006

Sept. 30, 2006

Hi all, I tried to compile scipy with Python 2.5 and Intel C++/Fortran compiler. The first import failed: In [1]:from scipy import * ------------------------------------------------------------ File "[...]/lib/python2.5/site-packages/scipy/maxentropy/maxentropy.py", line 69 from __future__ import division <type 'exceptions.SyntaxError'>: from __future__ imports must occur at the beginning of the file That problem was easy to fix just by following the instructions of that message in 2 files: scipy/maxentropy/maxentropy.py scipy/maxentropy/maxentutils.py Is that import necessary with Python 2.5 or even Python 2.4? Best regards, Lars

3 3

A question about the moin site
by Fernando Perez Sept. 28, 2006

Sept. 28, 2006

Hi all, we've just 'officially moved' ipython to a moin site similar to the scipy one, but I have a few questoins our site gurus may help with: 1. Figures: the scipy pages use for figures code like: {{{ #!figure ... }}} I tried that but it doesn't work out of the box. Is there something I can do on my side, or install in the wiki/ directory, for this to work? I'm sure it's something very simple, but googling hasn't revealed the magical incantation so far. 2. reST: is there an easy way to have moin link to reST documents and render them into HTML? Some of the new ipython docs (README, etc) are in reST and it would be great to show them nicely formatted on the site. 3. Latex: this parser seems very nice: http://johannes.sipsolutions.net/Projects/new-moinmoin-latex I'm currently testing it on the ipython moin, but perhaps the python process hasn't been reloaded, because so far it doesn't seem to work yet. I'll give it a bit more time (is there any signal one can send to some process to force a reload of moin when changes like this are being tested?). I was thinking it would be very nice to have reST and Latex support on the Scipy wiki as well, as I hope that in the long term, the scipy wiki should grow pages with detailed algorithmic and mathematical discussions, and for such, latex is necessary. Thanks for any help! Best, f

3 3

packages inside and outside scipy distribution
by David Cournapeau Sept. 28, 2006

Sept. 28, 2006

Hi, I would like to know what is the preferred approach to format a package for numpy/scipy, developed outside scipy (but eventually to be included in scipy), particularly the meta-files. I have already a working package, but I would like it to conform to scipy 'standards' for packaging (doc, tests, etc...). For now, the structure of the package is: root_dir/ where there are files such as README, LICENSE, Changelog, etc... root_dir/pkg_name where the implementation is (eventually in sub-directories) For now, the setup.py is in root_dir, because I don't want to 'pollute' root_dir/pkg_name with files which are not a part of the "binary" distribution (by binary, I mean here all the files required by the package to work, including the doc). If I understand correctly, a scipy package should not have a root_dir, but everything, including the setup.py in the root_dir/pkg_name. Is there a way to do that without polluting the directory with files produced by setup.py, such as dist dir, MANIFEST, etc... ? More fundamentally, what would be a sensible approach to develop a package 'outside scipy', with the goal to propose it for inclusion into scipy later ? cheers, David

1 0

Re: [SciPy-dev] patches for Intel MKL fft support (complex only)
by John Travers Sept. 22, 2006

Sept. 22, 2006

Thanks, I have done as you suggested. This topic can now be followed with scipy Trac ticket #268 Best regards, John On 20/09/06, Albert Strasheim <fullung(a)gmail.com> wrote: > Hey John > > Great patches! I suggest creating a new ticket in the SciPy Trac site so > that the SciPy developers can keep track of them. You can attach patches to > this ticket as you complete your work. > > Go here to do the necessary: > > 1. http://projects.scipy.org/scipy/scipy/register > 2. http://projects.scipy.org/scipy/scipy/login > 3. http://projects.scipy.org/scipy/scipy/newticket > > Regards, > > Albert > > > -----Original Message----- > > From: scipy-dev-bounces(a)scipy.org [mailto:scipy-dev-bounces@scipy.org] On > > Behalf Of John Travers > > Sent: 20 September 2006 19:25 > > To: scipy-dev(a)scipy.org > > Subject: [SciPy-dev] patches for Intel MKL fft support (complex only) > > > > Hi all, > > > > Attached are two patches that combined will enable support for the use > > of Intel MKL fft library calls in scipy.fftpack. At the moment only > > complex transforms zfft and zfftnd and their inverses are supported. > > Convolution and real transforms are in progress. > > > > <snup> > > -- My mobile (which I rarely answer): (+44) (0) 7739 105209 If you want to see what I do: www.femto.ph.ic.ac.uk

2 2

Include patch in ticket {1} (slow fft when using fftw3) ?
by David Cournapeau Sept. 22, 2006

Sept. 22, 2006

Hi, I would like to know if it would be possible to include the patch I posted to solve the slow fft in scipy problem (scipy ticket 1). I am using it myself for some time, never failed on me, and have a dramatic influence on speed (between 2 and 5 for complex input 1d fft), cheers, David

3 2

slice3.py and xplt
by Nils Wagner Sept. 21, 2006

Sept. 21, 2006

Hi all, Meanwhile I have replaced numpy.lib.mlab by numpy.oldnumeric.mlab in Mplot.py The nexr problem is >>> from scipy.sandbox import xplt Traceback (most recent call last): File "<stdin>", line 1, in ? File "/usr/lib64/python2.4/site-packages/scipy/sandbox/xplt/__init__.py", line 48, in ? from Mplot import * File "/usr/lib64/python2.4/site-packages/scipy/sandbox/xplt/Mplot.py", line 1013, in ? import colorbar File "/usr/lib64/python2.4/site-packages/scipy/sandbox/xplt/colorbar.py", line 11, in ? from slice3 import * File "/usr/lib64/python2.4/site-packages/scipy/sandbox/xplt/slice3.py", line 1585, in ? _poly_permutations4 = _construct3 (0) File "/usr/lib64/python2.4/site-packages/scipy/sandbox/xplt/slice3.py", line 1579, in _construct3 mask = find_mask (below, _node_edges3 [itype]) TypeError: array cannot be safely cast to required type How can I resolve this problem ? Nils

4 4