Mailman 3 January 2008 - SciPy-Dev

Nose tests - merged
by Matthew Brett Jan. 17, 2008

Jan. 17, 2008

I can now tell you that I have full knowledge of the problems of SVN and merging. I've just committed the nose testing modifications in a series of commits. Please let me know if there are any problems. Matthew

6 15

Re: [SciPy-dev] Sparse BLAS in scipy
by Anand Patil Jan. 16, 2008

Jan. 16, 2008

On Nov 28, 2007 4:06 PM, Nathan Bell <wnbell(a)gmail.com> wrote: > On Nov 27, 2007 4:35 PM, Anand Patil <anand.prabhakar.patil(a)gmail.com> > wrote: > > That and the analogous forward multiplication with a triangular matrix > are > > the two things I really need right now, yeah. The backend would be > simple, > > yet painful; upper/lower, transpose and side arguments, in addition to > > accounting for all the possible combinations of matrix formats, and the > > strides when 'B' is dense, would make for a huge number of bug > > opportunities. That's what made me think wrapping a library would be > less > > work than writing even 'two' routines from scratch. > > I guess I don't fully understand what you need here. What I imagined > was two methods for forward/backward solves with lower/upper > triangular matrices in CSR format. > I was thinking the triangular matrices could be any format, and the 'b' matrix/vector could be any format, in which case the multiplicity of functions rivals that of the ordinary multiplications involving sparse matrices. BTW, for such matrices, isn't this is equivalent to Gauss-Seidel[1] > sweep in the appropriate direction. If so, then we might just move > something like [2] to scipy.linsolve. > I can't see any difference... it would be nice to have the L3 version too, though, and in that case b could be in any format. Also, are you sure the sparse LU solvers wouldn't do this for you anyway? > Nope, will have to look more closely at those. > About a year ago I implemented sparsetools from scratch. I looked at > the same packages you listed before and came to the conclusion that > the amount of code needed was rather small and that by writing the > library to fit well with SciPy was worth the effort. > > Ultimately, I think making the backend code amenable to SWIG and > scipy.sparse saved more effort than starting with an existing library. OK, if you've done the research already and come to that conclusion I should just do it! Regardless of the backend (LU, gauss-seidel or from scratch) here's what I'm proposing for the Python interface: B=trimult(A, B, uplo='U', side='L', inplace=True) B=trisolve(A, B, uplo, side, inplace) A is triangular, B is rectangular or a vector. We could trim the uplo and side arguments if there were 'triangular matrix' classes that kept track of whether they're upper or lower, but that would be more trouble than it's worth, no? I'm thinking if inplace=True, B will only be overwritten if it's possible to recycle the memory efficiently, so the only safe usage is to use the return value. How does this sound? Anand

2 4

KeyError in scikits/audiolab
by Nils Wagner Jan. 16, 2008

Jan. 16, 2008

Hi all, The installation of scikits/audiolab also fails. Am I missing something ? /usr/bin/python setup.py install sndfile_info: libraries sndfile not found in /usr/local/lib FOUND: libraries = ['sndfile'] library_dirs = ['/usr/lib'] fulllibloc = /usr/lib/libsndfile.so.1 fullheadloc = /usr/include/sndfile.h include_dirs = ['/usr/include'] Traceback (most recent call last): File "setup.py", line 214, in ? classifiers = File "/usr/lib/python2.4/site-packages/numpy/distutils/core.py", line 147, in setup config = configuration() File "setup.py", line 158, in configuration from scikits.audiolab.info import VERSION as audiolab_version File "/home/nwagner/svn/audiolab/scikits/audiolab/__init__.py", line 24, in ? from pysndfile import formatinfo, sndfile File "/home/nwagner/svn/audiolab/scikits/audiolab/pysndfile.py", line 216, in ? py_to_snd_file_format_dic = { KeyError: 'SF_FORMAT_FLAC' Cheers, Nils

1 0

Trouble with scikits/ann
by Nils Wagner Jan. 16, 2008

Jan. 16, 2008

Hi all, I had some trouble with the installation of the scikits package ann. Traceback (most recent call last): File "setup.py", line 110, in ? test_suite = 'scikits.ann.tests.test_ann', File "/usr/lib/python2.4/site-packages/numpy/distutils/core.py", line 147, in setup config = configuration() File "setup.py", line 90, in configuration config.add_subpackage(DISTNAME) File "/usr/lib/python2.4/site-packages/numpy/distutils/misc_util.py", line 834, in add_subpackage caller_level = 2) File "/usr/lib/python2.4/site-packages/numpy/distutils/misc_util.py", line 817, in get_subpackage caller_level = caller_level + 1) File "/usr/lib/python2.4/site-packages/numpy/distutils/misc_util.py", line 749, in _get_configuration_from_setup_py ('.py', 'U', 1)) File "scikits/ann/setup.py", line 38, in ? ANN_ROOT = os.path.join(os.path.sep,*(__file__.split(os.path.sep)[-6])) IndexError: list index out of range How can I resolve the problem ? Cheers, Nils

1 0

Fedora Core 6 x86_64?
by Peter Skomoroch Jan. 16, 2008

Jan. 16, 2008

I'm having some trouble tracking down the best recipe for building an optimized numpy on FC6 x86_64 with BLAS/LAPACK. David's repository seems empty right now: http://software.opensuse.org/download/home:/ashigabou/ -Pete -- Peter N. Skomoroch peter.skomoroch(a)gmail.com http://www.datawrangling.com

1 0

Mac leopard and scipy sparse status
by John Ollinger Jan. 16, 2008

Jan. 16, 2008

I have been porting my numpy/scipy apps to my new imac for the last week or so , and have been getting seg faults in the sparse matrix package for the available binary build. I rebuilt lapack, atlas, umfpack and numpy from sources without problems, but the scipy build crashes while wrapping umfpack. I recall seeing some traffic about modifications to the sparse matrix package, and am wondering if I should get the svn version of this package before I starting working on the problem. It is probably just a change in the swig include file, but I use this package a lot and would like to start with a more recent version if there have been significant changes. John p.s. I would be happy to package up the release and contribute it if anyone is interested. Leopard ships gcc 4.0.2, which if I recall correctly, performed pretty badly on the atlas benchmarks. I installed 4.2.2 and am using that for the build, so it might be bit faster. -- John Ollinger University of Wisconsin Waisman Center, T233 1500 Highland Ave Madison, WI 53711 http://brainimaging.waisman.wisc.edu/~jjo/

2 3

sandbox multigrid
by Nils Wagner Jan. 15, 2008

Jan. 15, 2008

Hi Nathan, This is to let you know that your sandbox package multigrid comprises an ImportError I am using >>> scipy.__version__ '0.7.0.dev3837' ====================================================================== ERROR: Failure: ImportError (cannot import name augment_candidates) ---------------------------------------------------------------------- Traceback (most recent call last): File "/usr/lib/python2.4/site-packages/nose/loader.py", line 363, in loadTestsFromName module = self.importer.importFromPath( File "/usr/lib/python2.4/site-packages/nose/importer.py", line 39, in importFromPath return self.importFromDir(dir_path, fqname) File "/usr/lib/python2.4/site-packages/nose/importer.py", line 84, in importFromDir mod = load_module(part_fqname, fh, filename, desc) File "/usr/lib/python2.4/site-packages/scipy/sandbox/multigrid/tests/test_adaptive.py", line 8, in ? from scipy.sandbox.multigrid.adaptive import augment_candidates ImportError: cannot import name augment_candidates Cheers, Nils

1 0

slicing vs. advanced selection -- can be there smth in the middle? ; -)
by Yaroslav Halchenko Jan. 15, 2008

Jan. 15, 2008

Dear Scipy and Numpy gurus, Per our IRC coversation with tvaught I decided to expose my problem and wishes to the list. At the moment, there are 2 possibilities to select sub region of an array 1. slicing -- efficient memory vise -- no copying is done, it is just a view over the original data (thus even .flags.owndata=False). Needs to be done by specifying Slice instance(s) (explicitly or not) in the index, ie b=a[ 1:4, 2:5 ] 2. advanced selection where either a list of indexes is given or a mask c=a[ [1,2,3], [2,3,4] ] in this case the data gets copied In the application we are developing (pymvpa) we are dealing with relatively large arrays of data (primarily 2D while processing), where first dimension corresponds to different samples, 2nd -- to different features. The problems comes that we often need to sample from an array. For instance to check cross-validation on N-1 fold we are to generate N 'views' over original array. In each such "view" 1 sample (row) is not present while training, and it is used as a sample to test against later on. so at the end instead of N data records, in current implementation we end up with N*(N-1) records (if we are to keep those views for further analysis). But that is not only the case with the 1st dimension -- we have to do similarly 'evil' selection of the features, which again leads to quite a big waste of memory. Thus I wondered, is there any facility which could help us out (may be by sacrificing reasonable computation cost) and have really a view on top of an array. We don't really need a sparse representation -- we are selecting a set of rows and columns, so every column (and similar across rows) for a given 'view' uses the same steps/increments between the elements. I hope that my wording makes some sense ;-) If not -- please don't hesitate to buzz me to provide really a use case. Thank you in advance -- Yaroslav Halchenko Research Assistant, Psychology Department, Rutgers-Newark Student Ph.D. @ CS Dept. NJIT Office: (973) 353-5440x263 | FWD: 82823 | Fax: (973) 353-1171 101 Warren Str, Smith Hall, Rm 4-105, Newark NJ 07102 WWW: http://www.linkedin.com/in/yarik

2 2

Scipy server suffering again...
by Fernando Perez Jan. 14, 2008

Jan. 14, 2008

Howdy, I keep on getting, frequently, the by now familiar """Internal Server Error The server encountered an internal error or misconfiguration and was unable to complete your request. """ so doing anything around the site, using trac, moin, etc, is becoming rather difficult. I just noticed a load average on the box around 16, though no process is consuming any significant amount of CPU. If there's anything on our side (the individual project admins) we can do to help, please let us know. Cheers, f

5 11

adding a nice progressbar to scipy
by Ondrej Certik Jan. 12, 2008

Jan. 12, 2008

Hi, in almost all of my computational projects, I use this nice progressbar: http://pypi.python.org/pypi/progressbar/ it's very easy to use: widgets=["calculating", " ", progressbar.Percentage(), ' ', progressbar.Bar(), ' ', progressbar.ETA()] pbar=progressbar.ProgressBar(widgets=widgets,maxval=10000).start() for i in range(10000): pbar.update(i) #do some heavy calculation in each step pbar.finish() and it shows the progress, estimated time of arrival (ETA), it's completely configurable with many options, etc. I just asked the author and he made the code dual licensed under BSD and LGPL (the original license). Especially the ETA is very handy, because I get a clue how long I need to wait for my code to finish and I get it for free (see the above example). Would this be a good adition to scipy? This is an example of a small but handy code, that I am lazy to install, but if it was in scipy, I'd be much more easier to use it. I'd send a patch then and write a helper function, so that it can be used like this for users that just want the default behavior: pbar=progressbar(maxval=10000) for i in range(10000): pbar.update(i) #do some heavy calculation in each step pbar.finish() Ondrej

4 10