Mailman 3 October 2004 - NumPy-Discussion

Problems with complex matrices
by Nils Wagner 22 Oct '04

22 Oct '04

Hi all, Another bug is revealed Traceback (most recent call last): File "complex_it.py", line 6, in ? res=dot(A,x)-r File "/usr/lib/python2.3/site-packages/Numeric/dotblas/__init__.py", line 55, in dot if multiarray.array(a).shape == () or multiarray.array(b).shape == (): TypeError: a float is required Nils

1 0

Counting array elements
by Stephen Walton 21 Oct '04

21 Oct '04

Is there some simple way of counting the number of array elements which satisfy a certain condition? It is easy to do A[A<=1].sum() to sum all the values of A which are less than 1, but there doesn't seem to be a count() method. I tried (A<=1).sum() but this throws an exception at numarray 1.1. If I try sum(A<=value) I have to nest multiple sums if A has rank greater than 1, plus the sum overflows if A is large, apparently because boolean gets treated as Int8. (Try A=arange(1024,shape=(32,32));sum(sum(A<=1024)). You get zero.) The following works: array(A<=1024,type=Int32).sum() but is awkward. Am I missing an obvious better alternative? If not, I'm going to file an RFE :-) . -- Stephen Walton <stephen.walton(a)csun.edu> Dept. of Physics & Astronomy, Cal State Northridge

2 5

argmin and unsigned types
by Robert Kern 21 Oct '04

21 Oct '04

argmin locates the minimum by finding the maximum of the negative of the input. Unfortunately, for unsigned arrays, the negative has nothing to do with the actual numerical negative. Example: >>> from numarray import * >>> a = arange(10).astype(UInt8) >>> print a [0 1 2 3 4 5 6 7 8 9] >>> print -a [ 0 255 254 253 252 251 250 249 248 247] >>> argmin(a) 1 We need a separate argmin to handle these arrays properly. -- Robert Kern rkern(a)ucsd.edu "In the fields of hell where the grass grows high Are the graves of dreams allowed to die." -- Richard Harter

1 0

Re: numarray and ATLAS
by Sebastian Stark 21 Oct '04

21 Oct '04

> Perhaps this is a too recurrent subject, but I"m having problems when > making numarray to use ATLAS instead of the mini-lapack included. I had to change lapack_libs and lapack_dirs in addons.py to read: lapack_libs = ['lapack', 'f77blas', 'f2c', 'cblas', 'atlas', 'm'] lapack_dirs = ['/usr/local/lib/ATLAS'] I have all my .a files in /usr/local/lib/ATLAS so I can control which ones I'm actually linking against. mosel ~ % ls -l /usr/local/lib/ATLAS total 14608 -rw-r--r-- 1 root staff 7952316 Oct 20 10:03 libatlas.a -rw-r--r-- 1 root staff 277592 Oct 20 10:03 libcblas.a -rw-r--r-- 1 root staff 261060 Oct 20 10:45 libf2c.a -rw-r--r-- 1 root staff 353278 Oct 20 10:03 libf77blas.a -rw-r--r-- 1 root staff 5734736 Oct 20 10:42 liblapack.a -rw-r--r-- 1 root staff 324968 Oct 20 10:03 libtstatlas.a -Sebastian (and yes, I get a significant speed boost from ATLAS) -- Sebastian Stark -- http://www.kyb.tuebingen.mpg.de/~stark Max Planck Institute for Biological Cybernetics Spemannstr. 38, 72076 Tuebingen Phone: +49 7071 601 555 -- Fax: +49 7071 601 552

4 4

indexing on uninitialized arrays
by Sebastian Stark 21 Oct '04

21 Oct '04

In matlab I can do: >> x = [] x = [] >> x(2) = 1.4 x = 0 1.4000 >> x(2,4) = 2.9 x = 0 1.4000 0 0 0 0 0 2.9000 which means x expands as necessary depending on "how far" my indexing goes. Now I'm thinking about how to realize this with numarray. I could imagine to define a derived array type "SelfInflatingArray" which catches the IndexError exception and does the right thing then. Any better ideas? -Sebastian -- Sebastian Stark -- http://www.kyb.tuebingen.mpg.de/~stark Max Planck Institute for Biological Cybernetics Spemannstr. 38, 72076 Tuebingen Phone: +49 7071 601 555 -- Fax: +49 7071 601 552

1 0

Problems compiling numeric using python2.4 and VS.Net 2003
by Frank Lindseth 20 Oct '04

20 Oct '04

Hi, I need numeric in a python2.4 / win32 project. Is there a binary installer somewhere? I tried to compile it from source but ran into the following problem (se below): Where are the libs supposed to come from? - Frank C:\users\frankl\download\Numeric-23.5>c:\Python24\python.exe setup.py install running install running build running build_py running build_ext building 'lapack_lite' extension C:\Program Files\Microsoft Visual Studio .NET 2003\Vc7\bin\link.exe /DLL /nologo /INCREMENTAL:NO /LIBPATH:/usr/lib/atlas /LIBPATH:c:\Python24\libs /LIBPATH:c:\P ython24\PCBuild lapack.lib cblas.lib f77blas.lib atlas.lib g2c.lib /EXPORT:initl apack_lite build\temp.win32-2.4\Release\Src\lapack_litemodule.obj /OUT:build\lib .win32-2.4\lapack_lite.pyd /IMPLIB:build\temp.win32-2.4\Release\Src\lapack_lite. lib LINK : fatal error LNK1181: cannot open input file 'lapack.lib' error: command '"C:\Program Files\Microsoft Visual Studio .NET 2003\Vc7\bin\link .exe"' failed with exit status 1181 C:\users\frankl\download\Numeric-23.5>

4 6

numarray and ATLAS
by Francesc Alted 19 Oct '04

19 Oct '04

Hi, Perhaps this is a too recurrent subject, but I'm having problems when making numarray to use ATLAS instead of the mini-lapack included. I've installed ATLAS 3.6.0 on my pentium IV machine. I've made it a completely featured LAPACK by following the instructions in: http://math-atlas.sourceforge.net/errata.html#completelp and I'm pretty sure that the resulting library works. Now, after exporting USE_LAPACK and set the appropiate directory for lapack_dirs in addons.py, the compilation went well (however, I can see that lapack_litemodule.c is still being compiled, and I don't know if that's normal or not). The command I've used to install is: $ python setup.py install --gencode --home=/users/exp/alted/bin-i686 And the error that happens during the test phase follows: $ python Python 2.3.4 (#1, Jul 22 2004, 20:47:54) [GCC 3.3.2 20031022 (Red Hat Linux 3.3.2-1)] on linux2 Type "help", "copyright", "credits" or "license" for more information. >>> import numarray.testall as testall >>> testall.test() numarray: ((0, 1199), (0, 1199)) numarray.records: (0, 48) numarray.strings: (0, 176) numarray.memmap: (0, 82) numarray.objects: (0, 105) numarray.memorytest: (0, 16) numarray.examples.convolve: ((0, 20), (0, 20), (0, 20), (0, 20)) numarray.convolve: (0, 52) Traceback (most recent call last): File "<stdin>", line 1, in ? File "/users/exp/alted/bin-i686/lib/python/numarray/testall.py", line 24, in test result = eval(p+".test()") File "<string>", line 0, in ? File "/users/exp/alted/bin-i686/lib/python/numarray/fft/FFT.py", line 326, in test import dtest File "/users/exp/alted/bin-i686/lib/python/numarray/fft/dtest.py", line 238, in ? import numarray.random_array as random_array File "/users/exp/alted/bin-i686/lib/python/numarray/random_array/__init__.py", line 7, in ? from RandomArray2 import * File "/users/exp/alted/bin-i686/lib/python/numarray/random_array/RandomArray2.py", line 3, in ? import numarray.linear_algebra as linalg File "/users/exp/alted/bin-i686/lib/python/numarray/linear_algebra/__init__.py", line 1, in ? from LinearAlgebra2 import * File "/users/exp/alted/bin-i686/lib/python/numarray/linear_algebra/LinearAlgebra2.py", line 23, in ? import lapack_lite2 ImportError: /users/exp/alted/bin-i686/lib/python/numarray/linear_algebra/lapack_lite2.so: undefined symbol: dgesdd_ I've checked that dgesdd symbol exists on my liblapack.a: $ strings ~/bin-i686/lib/atlas/liblapack.a | grep dgesdd dgesdd.o/ 1097832195 2514 515 100644 13788 ` but not a dgesdd_, as you can see. I'm missing something? -- Francesc Alted

3 7

Numeric Underflow Exceptions: Recommendations?
by Frank Horowitz 19 Oct '04

19 Oct '04

Hi all, Using Numeric 23.5 I've been bitten by the dreaded 'floating point underflow throws an "OverflowError: math range error" instead of silently returning zero' bug. My setup is Debian unstable (Sid) on an i386, and I am using Debian's binary package "python-numeric". I understand from googling past discussions that this is (used to be?) phase-of-the-moon stuff, depending mostly upon architecture, options at libm compilation time of libc6. Several references to a trick of adding "-lieee" to the link list succeeding in taming the bug were mentioned around the era of Python2.0. My questions are these: Is there some higher level way of dealing with underflow now in Numeric? Or am I going to have to track down wherever "-lieee" has disappeared to in Debian, and recompile Numeric in the hopes that that still cures the problem? Any other tricks up people's sleeves for dealing with this? (I already know about exp_safe in Fernando Perez' IPython/numutils.py, BTW. I'm kind of hoping for a library level fix though, since my code is littered with "Numeric.exp()" calls.) TIA for any help you might be able to provide! Cheers, Frank Horowitz

1 0

New LAPACK and ScaLAPACK planned
by Stephen Walton 17 Oct '04

17 Oct '04

From volume 4 #37 of the NA-Digest mailing list. I hope this is of enough interest to this list to justify the cross post.

1 0

documentation error
by Alan G Isaac 16 Oct '04

16 Oct '04

In the Numeric manual, there are two different defintions of the 'diagonal' function. The second definition appears to be incorrect. p.39: diagonal(a, k=0, axis1=0, axis2 = 1) returns the entries along the k th diagonal of a (k is an offset from the main diagonal). This is designed for 2d arrays. For larger arrays, it will return the diagonal of each 2d sub-array. p.44 diagonal(a, offset=0, axis1=0, axis2=1) The diagonal function takes an array a, and returns an array of rank 1 containing all of the elements of a such that the difference between their indices along the specified axes is equal to the specified offset. With the default values, this corresponds to all of the elements of the diagonal of a along the last two axes. fwiw, Alan Isaac

4 8

NumPy-Discussion October 2004