Mailman 3 August 2006 - SciPy-User

Call for testing
by Nils Wagner Aug. 13, 2006

Aug. 13, 2006

Dear Scipy users, It's not my intention to annoy you, but is someone able to confirm the output (32bit.png) of test_det.py on a 32-bit machine ? The script is available via http://projects.scipy.org/scipy/scipy/ticket/223 I have no idea why I get different results on different architectures. Thanks in advance for testing. Nils

7 13

ODE solvers and Scipy
by JJ Aug. 13, 2006

Aug. 13, 2006

Hello. I am just about to start building a somewhat complex dynamic model that will incorporate a large number (~50-100) of differential equations and parameters to be estimated. This is my first project of this type, so I am on a steep learning curve. I could build it using Matlab's Simulink, but I would rather build it in Python. I have looked for good graphical interfaces to ODE solvers in Python but have not found any that seem suitable. I would not mind using the Scipy ODE solvers, but I have a couple of quick, simple questions before I invest a lot of time in that direction. Any thoughts would be appreciated. -- The (few) examples I have seen in using Scipy's ODE functions are only small, simple ones. Is there anything to prevent use of the solvers for larger, more complex systems of ODEs? -- Has anyone seen a good tutorial on use of Scipy's ODE solver for larger (or at least not tiny) problems? -- Are there other Open Source packages out there that might offer more options than Scipy's ODE solver? I dont want to make this job any harder than I need to. Ive looked but didnt see any. -- Simulink offers use of various model blocks to simplify model building. I can write my own, but is there any public code already available along these lines for use with Scipy's ODE solver? Please pardon the simplicity of my questions and thanks for your help. Would anyone be willing to allow me to write privately if I have a few troubles along the way? John __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com

5 5

Ubuntu speed-up
by Gary Ruben Aug. 13, 2006

Aug. 13, 2006

A friend sent me this link to a shell script which claims to speed up the standard Ubuntu Dapper Drake distro. I thought those here who quest for speed would be interested. Note; I haven't tried this yet but it looks safe: http://www.dylanknightrogers.com/2006/07/17/faster-dappersh/ Gary R

2 1

should "python setup.py sdist" provide the sandbox?
by Andrew Straw Aug. 12, 2006

Aug. 12, 2006

"python setup.py sdist" from the scipy svn tree doesn't grab the sandbox sources. I think it should. (People may want to play with the packages, for example.) A simple way to go about this would be to add all the files to MANIFEST.in. If folks think that's a good idea, I'll go ahead and generate a MANIFEST.in that adds all the files in svn scipy add submit the patch to Trac. Or is there some reason to exclude the source of the sandbox from "sdist" built distros?

3 4

fortran compiter, OS X 10.3
by John Hunter Aug. 12, 2006

Aug. 12, 2006

I am trying to compile scipy from svn 2156 on my powerbook OS X 10.3. I have gcc 3.3 installed but apparently no g77, g95 or f95. When I try and compile, I get the error since it can't find a fortran compiler. Before I go down a false path compiling a new gcc/g77, I was wondering if anybody had a recommendation for a binary g77 that I can get that is compatible with my gcc, or alternatively, what the path of least resistance is for me to get a compiler for this platform that can compile scipy. John-Hunters-Computer:~/python/svn/scipy> gcc --version gcc (GCC) 3.3 20030304 (Apple Computer, Inc. build 1495)

5 9

ODE solvers and Scipy
by John Aug. 12, 2006

Aug. 12, 2006

Thanks John and David. It looks as if I will try to use python/scipy, unless someone has written wrappers for Sundials for python. Does anybody know if this has been done? John

1 0

better control over odeint
by A. M. Archibald Aug. 12, 2006

Aug. 12, 2006

Hi, I'm writing a program that requires extensive use of scipy.integrate.odeint (tracing paths around a black hole). It mostly works well but has certain frustrating features. (1) It is apparently impossible to prevent it from printing to the standard output. This is wrong; if it must output error messages, they should go on standard error, so as not to contaminate the output of my program. It would also be nice to be able to turn them off! Or, more specifically, to signal the error conditions in a way I can reasonably test for. String matching on info["messages"] is not at all satisfactory. (2) It is very difficult to control what happens when an error occurs. When a function called from inside odeint throws an exception (undefined variable, overflow error, what have you), odeint catches it, prints a backtrace (to standard error, at least, thankfully), and continues trying more points. Since what has usually happened is that odeint has run into a singularity, it grinds away endlessly, spewing page after page of the same error message, before finally giving up. An error is then signaled by leaving info["mused"][n]==0. Ideal for my purposes would be for odeint simply not to catch exceptions, or to treat exceptions as an instruction to return early. Failure to integrate over the whole range of t could be signaled by raising an exception. This would allow easy catching of integration failures, and it would also (for example) allow me to signal that the ray had proceeded into a forbidden region by throwing an exception. I realize that nobody is very keen on rooting around inside the FORTRAN code for odeint, and I can't see directly how to make it stop immediately. It should be fine for the user function to simply return NaNs. Another option is, if the integration falls apart at some point, to simply stop and provide information about how far the integration successfully progressed. lsoda.f is definitely capable of this without modification. Thanks. A. M. Archibald

1 0

scipy.fftpack.fft2 bug
by Brice Thurin Aug. 11, 2006

Aug. 11, 2006

It seems there is a bug with the axes argument in the scipy.fftpack.fft2 function. It does not do what it is supose to do. In [1]: import scipy.fftpack In [2]: import numpy In [3]: x = ones((4,4,2)) In [4]: fx = scipy.fftpack.fft2(x,shape=(8,8),axes = (-3,-2)) In [5]: fx.shape Out[5]: (4, 8, 8) but it is working fine with numpy.dft.fft2 In [1]: import numpy In [2]: x = numpy.ones((4,4,2)) In [3]: fx = numpy.dft.fft2(x,s=(8,8),axes=(-3,-2)) In [4]: fx.shape Out[4]: (8, 8, 2)

1 0

Re: [SciPy-user] SciPy installation problems on SUSE linux
by Eric Emsellem Aug. 10, 2006

Aug. 10, 2006

Hi Sebastian, here is a (attached) file I used to install all the necessary packages; Things are much easier via svn of course. Hope this helps. Don't forget to remove the old stuff if you wish to reinstall scipy and numpy. Eric P.S.: I have Suse 10.1, and it installs fine on my laptop and desktop PC ####################### # LAPACK # ##################### # Build Lapack from the web since then you are sure it is complete # It is in the Suse distribution but I don't trust it wget http://www.netlib.org/lapack/lapack.tgz mkdir install_lapack cd install_lapack tar xvfz ../lapack.tgz cd LAPACK cp INSTALL/make.inc.LINUX make.inc ## This will build a lapack_LINUX.a ## That you will use later on for ATLAS make lapacklib ######################## # ATLAS # ###################### tar xvfz atlas3.6.0.tar.gz cd ATLAS ## You will have to answer a few questions... make # change "arch" with the right one # you will be asked a few question there make install arch=Linux_UNKNOWNSSE2 ## Create a new directory where to put the include and lib files ## (root) mkdir /usr/local/lib/atlas ## then copy all the include and archive (*.a) files ## under this new dir cp include/* /usr/local/lib/atlas cp *.a /usr/local/lib/atlas # in .cshrc # and don't forget to set it up setenv ATLAS /usr/local/lib/atlas ## save the lapack lib built from ATLAS (which is incomplete) cd /usr/local/lib/atlas cp liblapack.a liblapack.a_ATLAS # then create a new temporary dir which will allow you to complete # lapack mkdir tmp cd tmp # detar the atlas lapack lib ar x ../liblapack.a # then cp the true lapack linux lib into the liblapack.a file cp /soft/python/tar/install_lapack/LAPACK/lapack_LINUX.a ../liblapack.a # and add the files from the atlas lib # you will then have a full lapack lib with the atlas files too ar r ../liblapack.a *.o cd .. # remove the temp dir rm -rf tmp ## copy all *.a into your local atlas dir cp *.a /usr/local/lib/atlas # ###################### ## Added all fftw libraries from yast # ###################### # ###################### ## Adding pyfits # ###################### svn co http://astropy.scipy.org/svn/pyfits/trunk pyfits cd pyfits python setup.py install ########################################################### # NUMPY # ######### svn co http://svn.scipy.org/svn/numpy/trunk numpy cd numpy python setup.py install # ###################### ## INSTALLED IPYTHON # ###################### tar xvfz ipython-0.7.2.tar.gz cd ipython-0.7.2 python setup.py install # ###################### ## F2PY # ###################### (CVS password: guest) cvs -d :pserver:anonymous@cens.ioc.ee:/home/cvs login cvs -z6 -d :pserver:anonymous@cens.ioc.ee:/home/cvs checkout f2py2e cd f2py2e make install # ###################### ## SCIPY # ###################### ## Remove any scipy in your python before.. ## ALSO REMOVE the "build" directories from the "core" and "scipy" directories before ## doing the install!! svn co http://svn.scipy.org/svn/scipy/trunk scipy cd scipy python setup.py install ## Test:: cd # Start python and run as tests: # (level 10 will take a while ...) import scipy scipy.test(1,verbosity=10) scipy.test(10,verbosity=10) ################################# # Matplotlib ################################# svn co https://svn.sourceforge.net/svnroot/matplotlib/trunk/matplotlib matplotlib # Note that some matplotlibrc options have changed # such that tick ==> xtick + ytick ## not needed anymore ### # add one line in setupext.py: ### ### o.tcl_inc="/usr/include/" ### ### # before ### if not os.path.exists(o.tcl_inc): ### print 'cannot find tcl/tk headers. giving up.' ### sys.exit() ### return o # and run the usual python setup.py install

1 0

SciPy installation problems on SUSE linux
by Sebastian Busch Aug. 10, 2006

Aug. 10, 2006

Dear list, I do have problems installing SciPy on my computer. In short, when I think I would be finished and run " import scipy print "hallo" " it says: " Overwriting info=<function info at 0x406a2144> from scipy.misc (was <function info at 0x404bdd14> from numpy.lib.utils) hallo " My computer is a > uname -a > Linux e13-v1 2.6.13-15.11-smp #1 SMP Mon Jul 17 09:43:01 UTC 2006 i686 i686 i386 GNU/Linux Unfortunately, I did so much stuff that I cannot reconstruct what I did. However, I tried to stick to http://pong.tamu.edu/tiki/tiki-view_blog_post.php?blogId=6&postId=97 Installing NumPy seemed to work, importing it in a script gives no error. I then installed blas and lapack (and now even atlas) and after that scipy via python setup.py build sudo python setup.py install Perhaps someone had a similar problem and can halp? I would highly appreciate that. Thanks in advance, Sebastian.

2 1