Mailman 3 April 2006 - SciPy-User

Vectorize wrapped Fortran routine
by steffen.loeck＠gmx.de April 7, 2006

April 7, 2006

Hi, I would like to vectorize a Fortran routine wrapped with f2py using scipy.vectorize. With the old scipy this works fine but with the new one i get the following error: TypeError: object is not a callable Python object Wrapping was done with: f2py -m hermite -h hermite.pyf hermite.f f2py2.3 -c hermite.pyf hermite.f The routine works without using vectorize but scipy.vectorize(hermite.routine) fails. Is there any way to get this working under new scipy? Thanks, Steffen

2 1

iterating through permutations
by Jonathan Taylor April 7, 2006

April 7, 2006

just wondering -- is there any easy way to iterate over all permutations of, say, K integers in scipy? i know the package probstat does this, just wondered if it existed in scipy..... thanks, jonathan -- ------------------------------------------------------------------------ I'm part of the Team in Training: please support our efforts for the Leukemia and Lymphoma Society! http://www.active.com/donate/tntsvmb/tntsvmbJTaylor GO TEAM !!! ------------------------------------------------------------------------ Jonathan Taylor Tel: 650.723.9230 Dept. of Statistics Fax: 650.725.8977 Sequoia Hall, 137 www-stat.stanford.edu/~jtaylo 390 Serra Mall Stanford, CA 94305

2 1

scipy.io.write_array and the precision
by jaonary＠free.fr April 6, 2006

April 6, 2006

Hi all, I'm trying to write an ascii output of my computation. To do this I'm planing to use the package io of scipy. My problem is that I can't get a simple ascii output. The read_array function write the numbers in the scientific form : 110.330e10. And me, I'd like to write my number simply as 122.00323123. So how can I do the with this module ? Is it possible or I have to do this by hand ? Best regards

4 6

Conditional maximum entropy models
by Ed Schofield April 6, 2006

April 6, 2006

Hi all, I've just merged new functionality for conditional maximum entropy models into the SVN trunk. There are two small example scripts with comments that demonstrate how to use it, and docstrings that describe the class and its methods in more detail. Thanks to Matt Cooper for his feedback on getting it working! -- Ed

1 0

FAIL tests, seems fftpack related, check_dot takes forever.
by iCy-fLaME April 6, 2006

April 6, 2006

Problem: check_dot (scipy.lib.tests.test_blas.test_fblas1_simple) seems to take forever on my system, been running at 100% CPU time for at least five minutes on this P4 3GHz machine, I cancel the test at the end. Several FAIL tests related to fftpack, not sure they are important or not. Can some one point me a direction to resolve the problems or indeed is there a problem to begin with? OS: Fedora Core 4 Platform: i686 (smp sse sse2) Installed package from rpm repo: atlas-3.6.0-9.fc4 atlas-sse-3.6.0-9.fc4 atlas-sse2-3.6.0-9.fc4 atlas-3dnow-3.6.0-9.fc4 fftw-3.1-3.fc4 fftw-devel-3.1-3.fc4 fftw2-2.1.5-11.fc4 fftw2-devel-2.1.5-11.fc4 python-2.4.1-2 Installed from source: numpy 0.9.6 scipy 0.4.8 Failed test log: check_definition (scipy.fftpack.tests.test_pseudo_diffs.test_hilbert) ... FAIL check_random_even (scipy.fftpack.tests.test_pseudo_diffs.test_hilbert) ... FAIL check_tilbert_relation (scipy.fftpack.tests.test_pseudo_diffs.test_hilbert) ... FAIL check_definition (scipy.fftpack.tests.test_pseudo_diffs.test_ihilbert) ... FAIL check_itilbert_relation (scipy.fftpack.tests.test_pseudo_diffs.test_ihilbert) ... FAIL check_definition (scipy.fftpack.tests.test_pseudo_diffs.test_itilbert) ... FAIL check_definition (scipy.fftpack.tests.test_pseudo_diffs.test_shift) ... FAIL check_definition (scipy.fftpack.tests.test_pseudo_diffs.test_tilbert) ... FAIL check_random_even (scipy.fftpack.tests.test_pseudo_diffs.test_tilbert) ... FAIL line-search Newton conjugate gradient optimization routine ... ERROR check_definition (scipy.fftpack.tests.test_basic.test_fft) ... FAIL check_djbfft (scipy.fftpack.tests.test_basic.test_fft) ... FAIL check_definition (scipy.fftpack.tests.test_basic.test_fftn) ... FAIL check_definition (scipy.fftpack.tests.test_basic.test_ifft) ... FAIL check_djbfft (scipy.fftpack.tests.test_basic.test_ifft) ... FAIL check_random_complex (scipy.fftpack.tests.test_basic.test_ifft) ... FAIL check_random_real (scipy.fftpack.tests.test_basic.test_ifft) ... FAIL check_definition (scipy.fftpack.tests.test_basic.test_ifftn) ... FAIL check_definition (scipy.fftpack.tests.test_basic.test_irfft) ... FAIL check_dot (scipy.lib.tests.test_blas.test_fblas1_simple) Traceback (most recent call last): File "<stdin>", line 1, in ? File "/usr/lib/python2.4/site-packages/scipy/__init__.py", line 67, in test return ScipyTest(scipy).test(level, verbosity) File "/usr/lib/python2.4/site-packages/numpy/testing/numpytest.py", line 438, in test runner.run(all_tests) File "/usr/lib/python2.4/unittest.py", line 696, in run test(result) File "/usr/lib/python2.4/unittest.py", line 428, in __call__ return self.run(*args, **kwds) File "/usr/lib/python2.4/unittest.py", line 424, in run test(result) File "/usr/lib/python2.4/site-packages/numpy/testing/numpytest.py", line 139, in __call__ unittest.TestCase.__call__(self, result) File "/usr/lib/python2.4/unittest.py", line 281, in __call__ return self.run(*args, **kwds) File "/usr/lib/python2.4/unittest.py", line 260, in run testMethod() File "/usr/lib/python2.4/site-packages/scipy/lib/blas/tests/test_blas.py", line 76, in check_dot assert_almost_equal(f([3j,-4,3-4j],[2,3,1]),-9+2j) -- iCy-fLaME The body maybe wounded, but it is the mind that hurts.

1 0

Numpy on 64 bit Xeon with ifort and mkl
by Rahul Kanwar April 6, 2006

April 6, 2006

Hello, I am trying to compile Numpy on 64 bit Xeon with ifort and mkl libraries running Suse 10.0 linux. I had set the MKLROOT variable to the mkl library root but it could'nt find the 64 bit library. After a little bit of snooping I found the following in numpy/distutils/cpuinfo.py ------------------------------ def _is_XEON(self): return re.match(r'.*?XEON\b', self.info[0]['model name']) is not None _is_Xeon = _is_XEON ------------------------------ I changed XEON to Xeon and it worked and was able to indentify the em64t libraries. But it again got stuck with the following message. I used the following command to build Numpy python setup.py config_fc --fcompiler=intel install ------------------------------ building 'numpy.core._dotblas' extension compiling C sources gcc options: '-pthread -fno-strict-aliasing -DNDEBUG -O2 -fmessage-length=0 -Wall -D_FORTIFY_SOURCE=2 -g -fPIC' compile options: '-Inumpy/core/blasdot -I/opt/intel/mkl/8.0.2/include -Inumpy/core/include -Ibuild/src/numpy/core -Inumpy/core/src -Inumpy/core/include -I/usr/include/python2.4 -c' gcc -pthread -shared build/temp.linux-x86_64-2.4/numpy/core/blasdot/_dotblas.o -L/opt/intel/mkl/8.0.2/lib/em64t -lmkl_em64t -lmkl -lvml -lguide -lpthread -o build/lib.linux-x86_64-2.4/numpy/core/_dotblas.so /usr/lib64/gcc/x86_64-suse-linux/4.0.2/../../../../x86_64-suse-linux/bin/ld: /opt/intel/mkl/8.0.2/lib/em64t/libmkl_em64t.a(def_cgemm_omp.o): relocation R_X86_64_PC32 against `_mkl_blas_def_cgemm_276__par_loop0' can not be used when making a shared object; recompile with -fPIC /usr/lib64/gcc/x86_64-suse-linux/4.0.2/../../../../x86_64-suse-linux/bin/ld: final link failed: Bad value collect2: ld returned 1 exit status /usr/lib64/gcc/x86_64-suse-linux/4.0.2/../../../../x86_64-suse-linux/bin/ld: /opt/intel/mkl/8.0.2/lib/em64t/libmkl_em64t.a(def_cgemm_omp.o): relocation R_X86_64_PC32 against `_mkl_blas_def_cgemm_276__par_loop0' can not be used when making a shared object; recompile with -fPIC /usr/lib64/gcc/x86_64-suse-linux/4.0.2/../../../../x86_64-suse-linux/bin/ld: final link failed: Bad value collect2: ld returned 1 exit status error: Command "gcc -pthread -shared build/temp.linux-x86_64-2.4/numpy/core/blasdot/_dotblas.o -L/opt/intel/mkl/8.0.2/lib/em64t -lmkl_em64t -lmkl -lvml -lguide -lpthread -o build/lib.linux-x86_64-2.4/numpy/core/_dotblas.so" failed with exit status 1 ---------------------------------------------- i successfuly compiled it without the -lmkl_em64t flag but when i import numpy in python it gives error that some symbol is missing. I think that maybe if i use ifort as the linker instead ok gcc then things will work out properly, but i could'nt find how to change the linker to ifort. Aynone there who can help me with this problem ? regards, Rahul

1 0

Trouble Compiling SciPy 0.4.8 under Debian 3.1r0a on AMD64
by Gil Citro April 5, 2006

April 5, 2006

I'm trying to install SciPy 0.4.8 from source under Debian 3.1r0a on an AMD64 machine. When I type python setup.py install, I get this error /usr/bin/ld: /usr/local/lib/libfftw3.a(mapflags.o): relocation R_X86_64_32 can not be used when making a shared object; recompile with -fPIC /usr/local/lib/libfftw3.a: could not read symbols: Bad value collect2: ld returned 1 exit status /usr/bin/ld: /usr/local/lib/libfftw3.a(mapflags.o): relocation R_X86_64_32 can not be used when making a shared object; recompile with -fPIC /usr/local/lib/libfftw3.a: could not read symbols: Bad value collect2: ld returned 1 exit status error: Command "/usr/bin/g77 -shared build/temp.linux-x86_64-2.3/build/src/Lib/fftpack/_fftpackmodule.o build/temp.linux-x86_64-2.3/Lib/fftpack/src/zfft.o build/temp.linux-x86_64-2.3/Lib/fftpack/src/drfft.o build/temp.linux-x86_64-2.3/Lib/fftpack/src/zrfft.o build/temp.linux-x86_64-2.3/Lib/fftpack/src/zfftnd.o build/temp.linux-x86_64-2.3/build/src/fortranobject.o -L/usr/local/lib -Lbuild/temp.linux-x86_64-2.3 -ldfftpack -lfftw3 -lg2c-pic -o build/lib.linux-x86_64-2.3/scipy/fftpack/_fftpack.so" failed with exit status 1 FFTW 3.1.1 is installed from sources and tests good with make check. Has anyone seen this problem? The closest match I came across was this, but it doesn't suggest to me how to fix the problem. http://www.mail-archive.com/cross-lfs@linuxfromscratch.org/msg00411.html If anyone has a suggestion about what to try I'd appreciate it. Thanks. Gil

2 4

Re: [SciPy-user] [SciPy-dev] Statistics review months
by Webb Sprague April 5, 2006

April 5, 2006

I just wanted to add my two cents to the how should we incorporate statistics into scipy question. Take it with all the grains of salt necessary.... I might be able to play a statistician on TV, but not much more... Point 1: One of the *nicest* things about R (and I suppose S-Plus) is that when you do a statistical procedure, even a simple regression, it calculates all sorts of useful stuff for you, from diagnostic plots to all that fancy ANOVA stuff to the residuals etc, and stores it in an object that you can manipulate later. I am not sure how to do this "pythonically" in scipy, but I think it should be considered when we design a stats extension. Point 2: One of the most *annoying* things about R/S-Plus is that it assumes a user-interaction paradigm rather than a server paradigm. I think that as we conceptualize a scipy extension we should remember that a lot of people might want to do something like make statistical analyses available through a web interface (one reason for using for going with Python for me, anyway) and relying on a X system, on the fly generated graphics, a persistent session across commands, etc, would make that extremely difficult (I tried...). At the very least, I would need to be able to pickle the analysis results and save them in a database, retrieving both data and graphics (or nicely re-generating the graphics) later. Anyway, I just wanted to get that off my chest. I hope it makes the world a better place :)

2 1

Interpolation and Extrapolation
by Jose Luis Gomez Dans April 5, 2006

April 5, 2006

Hi! I want to interpolate and smooth a regular time series, with gaps. Gaps in the middle of the series are not a problem, as I want to have a periodic temporal sampling, and I can interpolate those points. However, I cannot extrapolate points which are in the extrema using the interpolate package (doh!). A first step is to use the nearest neighbour, but what´s the most efficient way of doing this? Is there a way of flagging the gaps (at the moment, they are set to zero), and using maybe interpolate.interp1d to get a first estimate of the time series prior to smoothing? Many thanks! Jose -- Analog-/ISDN-Nutzer sparen mit GMX SmartSurfer bis zu 70%! Kostenlos downloaden: http://www.gmx.net/de/go/smartsurfer

1 0

implementation for sine transformation?
by Christian Meesters April 5, 2006

April 5, 2006

Hi, I'm looking for a way of calculating a sine transformation in Python, which I'd like to apply on a 3D-array. Does anybody know an (tested) implementation? Or could somebody give me a hint how to achieve this using a function from fftpack as a shortcut? If nothing is already available: Would it be worth to implement it in SciPy? In this case I would try to code it with this goal in mind. (But I would need some time, since other projects have a certain priority these weeks.) Cheers Christian

3 3