Mailman 3 May 2010 - NumPy-Discussion

Re: [Numpy-discussion] Installing audiolab on python 2.6
by arthur de conihout 29 May '10

29 May '10

Hi I reinstall it on a different machine i got the same error with the nsdfile no warning anymore with audiolab . I followed these points: Requirements¶<http://www.ar.media.kyoto-u.ac.jp/members/david/softwares/audiolab/sphinx/i…> audiolab requires the following softwares: - a python interpreter. - libsndfile - numpy (any version >= 1.2 should work). - setuptools On Ubuntu, you can install the dependencies as follow: sudo apt-get install python-dev python-numpy python-setuptools libsndfile-dev Optional¶<http://www.ar.media.kyoto-u.ac.jp/members/david/softwares/audiolab/sphinx/i…> Audiolab can optionally install audio backends. For now, only alsa (Linux) and Core Audio (Mac OS X) are supported. On Linux, you need alsa headers for this to work; on Ubuntu, you can install them with the following command: sudo apt-get install libasound2-dev For Mac OS X, you need the CoreAudio framework, available on the Apple website. Build¶<http://www.ar.media.kyoto-u.ac.jp/members/david/softwares/audiolab/sphinx/i…>For unix users, if libsndfile is installed in standart location (eg /usr/lib, /usr/local/lib), the installer should be able to find them automatically, and you only need to do a “python setup.py install” Concerning the eventual site.cfg file i have no idea what to do with this it might be the problem maybe you can give advice? Thank you 2010/5/28 David Cournapeau <cournape(a)gmail.com> > On Fri, May 28, 2010 at 1:58 AM, arthur de conihout > <arthurdeconihout(a)gmail.com> wrote: > > Thank you for the answer > > i would love trying audiolab but i got this error while importing i m > > running python2.6 > > > >>>> import audiolab > > > /usr/local/lib/python2.6/dist-packages/audiolab-0.0.0-py2.6-linux-i686.egg/audiolab/soundio/play.py:48: > > UserWarning: Could not import alsa backend; most probably, you did not > have > > alsa headers when building audiolab > > How did you install it ? Your installation looks bogus (if you build > it by yourself, please give the *exact* instructions you used) > > David > _______________________________________________ > NumPy-Discussion mailing list > NumPy-Discussion(a)scipy.org > http://mail.scipy.org/mailman/listinfo/numpy-discussion >

1 1

curious about how people would feel about moving to github
by Jarrod Millman 28 May '10

28 May '10

Hello, I changed the subject line for this thread, since I didn't want to hijack another thread. Anyway, I am not proposing that we actually decide whether to move to git and github now, but I am just curious how people would feel. We had a conversation about this a few years ago and it was quite contentious at the time. Since then, I believe a number of us have started using git and github for most of our work. And there are a number of developers using git-svn to develop numpy now. So I was curious to get a feeling for what people would think about it, if we moved to git. (I don't want to rehash the arguments for the move.) Anyway, Chuck listed the main concerns we had previously when we discussed moving from svn to git. See the discussion below. Are there any other concerns? Am I right in thinking that most of the developers would prefer git at this point? Or are there still a number of developers who would prefer using svn still? On Wed, May 26, 2010 at 12:54 PM, Charles R Harris <charlesr.harris(a)gmail.com> wrote: > I think the main problem has been windows compatibility. Git is best from > the command line whereas the windows command line is an afterthought. > Another box that needs a check-mark is the buildbot. If svn clients are > supported then it may be that neither of those are going to be a problem. I was under the impression that there were a number of decent git clients for Windows now, but I don't know anyone who develops on Windows. Are there any NumPy developers who use Windows who could check out the current situation? Pulling from github with an svn client works very well, so buildbot could continue working as is: http://github.com/blog/626-announcing-svn-support And if it turns out the Windows clients are still not good enough, we could look into the recently add svn write support to github: http://github.com/blog/644-subversion-write-support No need for us to make any changes immediately. I am just curious how people would feel about it at this point. Jarrod

14 41

crash in np.poly
by josef.pktd＠gmail.com 28 May '10

28 May '10

I was tracking a test failure/crash in scipy.signal.ltisys numpy 1.4.0: >>> np.poly(np.zeros((0,0))) ** On entry to DGEEV parameter number 5 had an illegal value Josef

2 3

Re: [Numpy-discussion] Help Convolution with binaural filters(HRTFs)
by arthur de conihout 28 May '10

28 May '10

> > > > Hi, >> i try to implement a real-time convolution module refreshed by head >> listener location (angle from a reference point).The result of the >> convolution by binaural flters(HRTFs) allows me to spatialize a monophonic >> wavfile. I got trouble with this as long as my convolution doesnt seem to >> work properly: >> np.convolve() doesnt convolve the entire entry signal >> >> ->trouble with extracting numpyarrays from the audio wav. filters and >> monophonic entry >> ->trouble then with encaspulating the resulting array in a proper wav >> file...it is not read by audacity >> >> Do you have any idea of how this could work or do you have any >> implementation of stereo filtering by impulse response to submit me >> Thank you very much >> >> Arthur de Conihout >> >> > > > >

3 9

NotImplemented returns
by josef.pktd＠gmail.com 28 May '10

28 May '10

A while ago we had a brief discussion about this. Is this a feature? or should there be a ticket for this >>> np.sqrt('5') NotImplemented >>> a = np.sqrt('5') >>> a NotImplemented >>> type(a) <type 'NotImplementedType'> Josef

3 5

StarCluster 0.91 - NumPy/SciPy Clusters on EC2
by Justin Riley 28 May '10

28 May '10

This is a one-time message to announce the availability of version 0.91 of the StarCluster package. Why should you care? StarCluster allows you to create NumPy/SciPy clusters configured with NFS-shared filesystems and the Sun Grid Engine queueing system out of the box on Amazon's Elastic Compute Cloud (EC2). The NumPy/SciPy installations have been compiled against a custom-compiled ATLAS for the larger EC2 instances. About ----- There is an article about StarCluster on www.hpcinthecloud.com: http://www.hpcinthecloud.com/features/StarCluster-Brings-HPC-to-the-Amazon-… There is also a screencast of installing, configuring, launching, and terminating an HPC cluster on Amazon EC2: http://www.hpcinthecloud.com/blogs/MITs-StarCluster-An-Update-with-Screenca… Project description from PyPI: StarCluster is a utility for creating and managing scientific computing clusters hosted on Amazon's Elastic Compute Cloud (EC2). StarCluster utilizes Amazon's EC2 web service to create and destroy clusters of Linux virtual machines on demand. To get started, the user creates a simple configuration file with their AWS account details and a few cluster preferences (e.g. number of machines, machine type, ssh keypairs, etc). After creating the configuration file and running StarCluster's "start" command, a cluster of Linux machines configured with the Sun Grid Engine queuing system, an NFS-shared /home directory, and OpenMPI with password-less ssh is created and ready to go out-of-the-box. Running StarCluster's "stop" command will shutdown the cluster and stop paying for service. This allows the user to only pay for what they use. StarCluster provides a Ubuntu-based Amazon Machine Image (AMI) in 32bit and 64bit architectures. The AMI contains an optimized NumPy/SciPy/Atlas/Blas/Lapack installation compiled for the larger Amazon EC2 instance types. The AMI also comes with Sun Grid Engine (SGE) and OpenMPI compiled with SGE support. The public AMI can easily be customized by launching a single instance of the public AMI, installing additional software on the instance, and then using StarCluster can also utilize Amazon's Elastic Block Storage (EBS) volumes to provide persistent data storage for a cluster. EBS volumes allow you to store large amounts of data in the Amazon cloud and are also easy to back-up and replicate in the cloud. StarCluster will mount and NFS-share any volumes specified in the config. StarCluster's "createvolume" command provides the ability to automatically create, format, and partition new EBS volumes for use with StarCluster. Download -------- StarCluster is available on PyPI (http://pypi.python.org/pypi/StarCluster) and also on the project's website: http://web.mit.edu/starcluster You will find the docs as well as links to the StarCluster mailing list on the website. New in this version: -------------------- * support for launching and managing multiple clusters on EC2 * added "listclusters" command for showing all active clusters on EC2 * support for attaching and NFS-sharing multiple EBS volumes * added createimage and createvolume commands for easily creating new AMIs and EBS volumes for use with StarCluster * experimental support for launching clusters using spot instances * added support for StarCluster "plugins" that provide the ability to perform additional configuration/setup routines on top of StarCluster's default cluster configuration * added "listpublic" command for listing all available public StarCluser AMIs that can be used with StarCluster * bash/zsh command line completion for StarCluster's command line interface

1 0

error during config on Ubuntu powerpc64, numpy-1.4.1
by Chris Kees 28 May '10

28 May '10

Hi, I'm getting an error in check_long_double_representation on a linux/powerpc64 box. Has anybody seen this before/know a fix? -Chris > python -V Python 2.6.5 > uname -a Linux chl-29-200 2.6.32-21-powerpc64-smp #32-Ubuntu SMP Fri Apr 16 10:28:57 UTC 2010 ppc64 GNU/Linux >python setup.py install ... C compiler: gcc -m32 -fno-strict-aliasing -DNDEBUG -g -O3 -Wall -Wstrict-prototypes -fPIC compile options: '-Inumpy/core/src/private -Inumpy/core/src -Inumpy/core -Inumpy/core/src/npymath -Inumpy/core/src/multiarray -Inumpy/core/src/umath -Inumpy/core/include -I/home/cekees/src/proteus/linux /include/python2.6 -c' gcc: _configtest.c removing: _configtest.c _configtest.o Traceback (most recent call last): File "setup.py", line 187, in <module> setup_package() File "setup.py", line 180, in setup_package configuration=configuration ) File "/home/cekees/src/proteus/externalPackages/numpy-1.4.1/numpy/distutils/core.py", line 186, in setup return old_setup(**new_attr) File "/home/cekees/src/proteus/linux/lib/python2.6/distutils/core.py", line 152, in setup dist.run_commands() File "/home/cekees/src/proteus/linux/lib/python2.6/distutils/dist.py", line 975, in run_commands self.run_command(cmd) File "/home/cekees/src/proteus/linux/lib/python2.6/distutils/dist.py", line 995, in run_command cmd_obj.run() File "/home/cekees/src/proteus/externalPackages/numpy-1.4.1/numpy/distutils/command/build.py", line 37, in run old_build.run(self) File "/home/cekees/src/proteus/linux/lib/python2.6/distutils/command/build.py", line 134, in run self.run_command(cmd_name) File "/home/cekees/src/proteus/linux/lib/python2.6/distutils/cmd.py", line 333, in run_command self.distribution.run_command(command) File "/home/cekees/src/proteus/linux/lib/python2.6/distutils/dist.py", line 995, in run_command cmd_obj.run() File "/home/cekees/src/proteus/externalPackages/numpy-1.4.1/numpy/distutils/command/build_src.py", line 152, in run self.build_sources() File "/home/cekees/src/proteus/externalPackages/numpy-1.4.1/numpy/distutils/command/build_src.py", line 169, in build_sources self.build_extension_sources(ext) File "/home/cekees/src/proteus/externalPackages/numpy-1.4.1/numpy/distutils/command/build_src.py", line 328, in build_extension_sources sources = self.generate_sources(sources, ext) File "/home/cekees/src/proteus/externalPackages/numpy-1.4.1/numpy/distutils/command/build_src.py", line 385, in generate_sources source = func(extension, build_dir) File "numpy/core/setup.py", line 413, in generate_config_h rep = check_long_double_representation(config_cmd) File "numpy/core/setup_common.py", line 136, in check_long_double_representation type = long_double_representation(pyod(object)) File "numpy/core/setup_common.py", line 244, in long_double_representation raise ValueError("Unrecognized format (%s)" % saw) ValueError: Unrecognized format (['001', '043', '105', '147', '211', '253', '315', '357', '301', '235', '157', '064', '124', '000', '000', '000', '000', '000', '000', '000', '000', '000', '000', '000', '376', '334', '272', '230', '166', '124', '062', '020'])

1 0

Extending documentation to c code
by Charles R Harris 28 May '10

28 May '10

Hi All, I'm wondering if we could extend the current documentation format to the c source code. The string blocks would be implemented something like /**NpyDoc """The Answer. Answer the Ultimate Question of Life, the Universe, and Everything. Parameters ---------- We don't need no stinkin' parameters. Notes ----- The run time of this routine may be excessive. """ */ int answer_ultimate_question(void) { return 42; } and the source scanned to generate the usual documentation. Thoughts? Chuck

7 20

How to distinguish between number and string dypes
by Vincent Davis 28 May '10

28 May '10

How do I determine if an array's (or column in a structured array) dtype is a number or a string. I see how to determine the actual dtype but all I want to know is if it is a string or a number. *Vincent Davis 720-301-3003 * vincent(a)vincentdavis.net my blog <http://vincentdavis.net> | LinkedIn<http://www.linkedin.com/in/vincentdavis>

6 8

zeros_like and friends shouldn't use ndarray.__new__(type(a), ...)
by Jim Porter 27 May '10

27 May '10

Ok, let's try sending this message again, since it looks like I can't send from gmane... (See discussion on python-list at http://permalink.gmane.org/gmane.comp.python.general/661328 for context) numpy.zeros_like contains the following code: def zeros_like(a): if isinstance(a, ndarray): res = ndarray.__new__(type(a), a.shape, a.dtype, order=a.flags.fnc) res.fill(0) return res ... This is a problem because basetype.__new__(subtype, ...) raises an exception when subtype is defined from C (specifically, when Py_TPFLAGS_HEAPTYPE is not set). There's a check in Objects/typeobject.c in tp_new_wrapper that disallows this (you can grep for "is not safe" to find there the exception is raised). The end result is that it's impossible to use zeros_like, ones_like or empty_like with ndarray subtypes defined in C. While I'm still not sure why Python needs this check in general, Robert Kern pointed out that the problem can be fixed pretty easily in NumPy by changing zeros_like and friends to something like this (with some modifications from me): def zeros_like(a): if isinstance(a, ndarray): res = numpy.zeros(a.shape, a.dtype, order=a.flags.fnc) res = res.view(type(a)) res.__array_finalize__(a) return res ... - Jim

2 1

NumPy-Discussion May 2010