From jose.martin at wanadoo.es  Sun Feb  1 20:28:47 2004
From: jose.martin at wanadoo.es (=?iso-8859-1?Q?Jos=E9_A_Mart=EDn_H?=)
Date: Mon, 2 Feb 2004 02:28:47 +0100
Subject: [SciPy-user] PYTHON23 WINDOWS BINARIES
Message-ID: <000f01c3e92b$e8d0a800$ba16243e@CASA>

IN ORDER TO USE THE SCIPY WITH PYTHON23 IN WIN32 FROM BINARY INSTALATION
YOU MUST INSTALL THE SCIPY CORE PACKAGE

I HAVE INSTALLED THIS, I DONT HAVE ANY GARRANTY THAT THIS PACKAGE DONT HAVE VIRUS BUT I HAVE INSTALLED AND SCIPY NOW WORK WELL.

THANKS TO GARY RUBEN...

http://users.bigpond.net.au/gazzar/Scipy_core-0.2.1_alpha_40.1444.win32-py23.exe

Please download this package and install it before using scipy binary installation for python 23.

Jose.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.scipy.org/pipermail/scipy-user/attachments/20040202/ea3e9e09/attachment.html>

From gazzar at email.com  Sun Feb  1 23:11:29 2004
From: gazzar at email.com (Gary Ruben)
Date: Mon, 02 Feb 2004 14:11:29 +1000
Subject: [SciPy-user] PYTHON23 WINDOWS BINARIES
Message-ID: <20040202041129.11254.qmail@email.com>

I noticed Jos? was having what seemed to be the same problem as Gey-Hong Gweon and I, so I made the scipy-core installer which Travis sent me offlist available for him to download. The simplest way to do this was to make it available on my site. However, I didn't intend this to be made public. Actually, I wouldn't mind if only a handful of people downloaded this, but after a few tens of people, my ISP will probably limit access, so it's not really practical to leave it there.
So, can whoever built the scipy Windows binary installer for Python 2.3 and included it on the download page please include scipy-core with it, or at least host the scipy-core installer on scipy.org and link to it from the downloads page.
thanks,
Gary

----- Original Message -----
From: Jos? A Mart?n H <jose.martin at wanadoo.es>
Date: Mon, 2 Feb 2004 02:28:47 +0100
To: "SciPy Users List" <scipy-user at scipy.net>
Subject: [SciPy-user] PYTHON23 WINDOWS BINARIES

> IN ORDER TO USE THE SCIPY WITH PYTHON23 IN WIN32 FROM BINARY INSTALATION
> YOU MUST INSTALL THE SCIPY CORE PACKAGE
> 
> I HAVE INSTALLED THIS, I DONT HAVE ANY GARRANTY THAT THIS PACKAGE DONT HAVE VIRUS BUT I HAVE INSTALLED AND SCIPY NOW WORK WELL.
> 
> THANKS TO GARY RUBEN...
> 
> http://users.bigpond.net.au/gazzar/Scipy_core-0.2.1_alpha_40.1444.win32-py23.exe
> 
> Please download this package and install it before using scipy binary installation for python 23.
> 
> Jose.

-- 
___________________________________________________________
Sign-up for Ads Free at Mail.com
http://promo.mail.com/adsfreejump.htm


From pearu at scipy.org  Mon Feb  2 17:54:24 2004
From: pearu at scipy.org (Pearu Peterson)
Date: Mon, 2 Feb 2004 16:54:24 -0600 (CST)
Subject: [SciPy-user] Scipy_core win32 installer available
In-Reply-To: <20040202041129.11254.qmail@email.com>
Message-ID: <Pine.LNX.4.44.0402021637030.6087-100000@scipy.org>


Hi,

You can find Scipy_core windows installer at my scipy page:

  http://cens.ioc.ee/~pearu/scipy/

Scipy_core requires that the following win32 software is installed:
  Python: 2.3.3, I got it from www.python.org
  Numeric: 23.1, I got it from www.numpy.org

After installing Scipy_core, recent Scipy win32 port 
from scipy.org should work (though I haven't tested this 
myself, yet).

Enjoy!
Pearu

PS: this is my first python extension built for win32 platform,
I shall celebrate this with a cup of tea..


From chris at fonnesbeck.org  Wed Feb  4 14:19:17 2004
From: chris at fonnesbeck.org (Christopher Fonnesbeck)
Date: Wed, 4 Feb 2004 14:19:17 -0500
Subject: [SciPy-user] distutils problem on cvs install under python 2.3
Message-ID: <06CCAB30-5747-11D8-838F-000A956FDAC0@fonnesbeck.org>

When I attempt a build of the latest cvs on a clean Debian Linux 
install of python2.3, I get this old error message:

chris at redhead:/usr/src/scipy$ python setup.py build
Traceback (most recent call last):
   File "setup.py", line 25, in ?
     from scipy_distutils.core import setup
   File "scipy_core/scipy_distutils/core.py", line 1, in ?
     from distutils.core import *
ImportError: No module named distutils.core


I thought this bug in distutils was squashed long ago. Am I missing 
something here?

Thanks,
C.
--
Christopher J. Fonnesbeck ( c h r i s @ f o n n e s b e c k . o r g )
Georgia Cooperative Fish & Wildlife Research Unit, University of Georgia


From pearu at scipy.org  Wed Feb  4 14:39:57 2004
From: pearu at scipy.org (Pearu Peterson)
Date: Wed, 4 Feb 2004 13:39:57 -0600 (CST)
Subject: [SciPy-user] distutils problem on cvs install under python 2.3
In-Reply-To: <06CCAB30-5747-11D8-838F-000A956FDAC0@fonnesbeck.org>
Message-ID: <Pine.LNX.4.44.0402041332090.9269-100000@scipy.org>



On Wed, 4 Feb 2004, Christopher Fonnesbeck wrote:

> When I attempt a build of the latest cvs on a clean Debian Linux 
> install of python2.3, I get this old error message:
> 
> chris at redhead:/usr/src/scipy$ python setup.py build
> Traceback (most recent call last):
>    File "setup.py", line 25, in ?
>      from scipy_distutils.core import setup
>    File "scipy_core/scipy_distutils/core.py", line 1, in ?
>      from distutils.core import *
> ImportError: No module named distutils.core

Could you verify that adding `import distutils` before the line #1
fixes this import error?

> I thought this bug in distutils was squashed long ago. Am I missing 
> something here?

Strange, I am on debian too with Python 2.3.3 and I don't get this error.
Btw, do you have write permissions to 
/usr/src/scipy/scipy_core/scipy_distutils directory? I noticed that
the traceback comes from .py file and not from .pyc file.

Pearu


From chris at fonnesbeck.org  Wed Feb  4 15:58:10 2004
From: chris at fonnesbeck.org (Christopher Fonnesbeck)
Date: Wed, 4 Feb 2004 15:58:10 -0500
Subject: [SciPy-user] distutils problem on cvs install under python 2.3
In-Reply-To: <Pine.LNX.4.44.0402041332090.9269-100000@scipy.org>
References: <Pine.LNX.4.44.0402041332090.9269-100000@scipy.org>
Message-ID: <D73F58E8-5754-11D8-838F-000A956FDAC0@fonnesbeck.org>


On Feb 4, 2004, at 2:39 PM, Pearu Peterson wrote:

>
>
> On Wed, 4 Feb 2004, Christopher Fonnesbeck wrote:
>
>> When I attempt a build of the latest cvs on a clean Debian Linux
>> install of python2.3, I get this old error message:
>>
>> chris at redhead:/usr/src/scipy$ python setup.py build
>> Traceback (most recent call last):
>>    File "setup.py", line 25, in ?
>>      from scipy_distutils.core import setup
>>    File "scipy_core/scipy_distutils/core.py", line 1, in ?
>>      from distutils.core import *
>> ImportError: No module named distutils.core
>
> Could you verify that adding `import distutils` before the line #1
> fixes this import error?

It does not! Very strange.


>
>> I thought this bug in distutils was squashed long ago. Am I missing
>> something here?
>
> Strange, I am on debian too with Python 2.3.3 and I don't get this 
> error.
> Btw, do you have write permissions to
> /usr/src/scipy/scipy_core/scipy_distutils directory? I noticed that
> the traceback comes from .py file and not from .pyc file.

I tried this running as root, so permissions should not be a problem.

C.

--
Christopher J. Fonnesbeck ( c h r i s @ f o n n e s b e c k . o r g )
Georgia Cooperative Fish & Wildlife Research Unit, University of Georgia


From ehaux at ucmerced.edu  Thu Feb  5 17:06:48 2004
From: ehaux at ucmerced.edu (ehaux at ucmerced.edu)
Date: Thu, 05 Feb 2004 14:06:48 -0800
Subject: [SciPy-user] Re: need help with gcc error( digest Vol 5, Issue 7)
Message-ID: <8282887c2b.87c2b82828@ucmerced.edu>

following up from Jan 27....

my system_info.py output is this:
atlas_info:
  FOUND:
    libraries = ['lapack', 'f77blas', 'cblas', 'atlas']
    library_dirs = ['/usr/lib/atlas', '/usr/lib']
 
blas_info:
  NOT AVAILABLE
 
blas_src_info:
  NOT AVAILABLE
 
dfftw_info:
  NOT AVAILABLE
 
dfftw_threads_info:
  NOT AVAILABLE
 
djbfft_info:
  NOT AVAILABLE
 
fftw_info:
  FOUND:
    libraries = ['rfftw', 'fftw']
    library_dirs = ['/usr/lib']
    define_macros = [('SCIPY_FFTW_H', None)]
    include_dirs = ['/usr/include']
 
fftw_threads_info:
  FOUND:
    libraries = ['rfftw_threads', 'fftw_threads']
    library_dirs = ['/usr/lib']
    define_macros = [('SCIPY_FFTW_THREADS_H', None)]
    include_dirs = ['/usr/include']
 
lapack_info:
  NOT AVAILABLE
 
lapack_src_info:
  NOT AVAILABLE
 
sfftw_info:
  NOT AVAILABLE
 
sfftw_threads_info:
  NOT AVAILABLE
 
x11_info:
  NOT AVAILABLE
 

I'm running Debian on a pIII and from $ uname -a :

Linux 2.6.1 #2 Tue Jan 20 15:33:43 PST 2004 i686 GNU/Linux

locate did not find Xlib.h

Eric

> Message: 1
> Date: Tue, 27 Jan 2004 01:35:02 -0600 (CST)
> From: Pearu Peterson <pearu at scipy.org>
> Subject: Re: [SciPy-user] need help with gcc error
> To: SciPy Users List <scipy-user at scipy.net>
> 
> > can anyone help with this??
> > i downloaded from cvs and did:
> > $python setup.py build
> > 
> > but got the following errors:
> > 
> > uild/temp.linux-i686-2.3/Lib/xplt/src/play/x11/colors.o -
> DGISTPATH="/usr/lib/python2.3/site-packages/scipy/xplt/gistdata"
> > In file included from Lib/xplt/src/play/x11/colors.c:9:
> > Lib/xplt/src/play/x11/playx.h:11:22: X11/Xlib.h: No such file or 
> directory                                      
> ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^What is the output of 
> 
>  python scipy_core/scipy_distutils/system_info.py
> 
> ? 
> Where the header file Xlib.h is located in your system?
> And could you give more details about your system 
> (sys.platform,os.name,uname -a, etc)?
> 
> Pearu



From ehaux at ucmerced.edu  Thu Feb  5 17:22:00 2004
From: ehaux at ucmerced.edu (ehaux at ucmerced.edu)
Date: Thu, 05 Feb 2004 14:22:00 -0800
Subject: [SciPy-user] Re: need help with gcc error (Digest, Vol 5, Issue 7)
Message-ID: <84d44854ec.854ec84d44@ucmerced.edu>

Also following up from Jan 27...

I get the following when using this script:

:/usr/lib/python2.3/site-packages/scipy$ python setup.py
/usr/bin/python scipy_core/setup.py
SciPy Core Version 0.2.1_alpha_40.1448
usage: setup.py [global_opts] cmd1 [cmd1_opts] [cmd2 [cmd2_opts] ...]
   or: setup.py --help [cmd1 cmd2 ...]
   or: setup.py --help-commands
   or: setup.py cmd --help
 
error: no commands supplied
Traceback (most recent call last):
  File "setup.py", line 127, in ?
    setup_package(ignore_packages)
  File "setup.py", line 77, in setup_package
    assert not s,'failed on scipy_core'
AssertionError: failed on scipy_core

> 
> Message: 3
> Date: Tue, 27 Jan 2004 06:17:31 -500
> From: scipy at zunzun.com (James R. Phillips)
> Subject: Re:  [SciPy-user] need help with gcc error
> To: scipy-user at scipy.net
> Message-ID: <401648cb9456c1.76981309 at mercury.sabren.com>
> Content-Type: text/plain; charset="iso-8859-1"
> 
> > i downloaded from cvs and did $python setup.py build
> > but got the following errors:
> > 
> > Lib/xplt/src/play/x11/playx.h:11:22: X11/Xlib.h: No such file or 
> directory
> You may need to configure the setup.py file for
> scipy to not build anything with X windows.  Here
> is the file I used as an example:
> 
> http://zunzun.com/no_X_setup.py
> 
>     James Phillips
> 



From pearu at scipy.org  Fri Feb  6 03:26:08 2004
From: pearu at scipy.org (Pearu Peterson)
Date: Fri, 6 Feb 2004 02:26:08 -0600 (CST)
Subject: [SciPy-user] Re: need help with gcc error( digest Vol 5, Issue
 7)
In-Reply-To: <8282887c2b.87c2b82828@ucmerced.edu>
Message-ID: <Pine.LNX.4.44.0402060216350.3227-100000@scipy.org>



On Thu, 5 Feb 2004 ehaux at ucmerced.edu wrote:

> following up from Jan 27....
> 
> my system_info.py output is this:

<snip>

> x11_info:
>   NOT AVAILABLE
>  
> 
> I'm running Debian on a pIII and from $ uname -a :
> 
> Linux 2.6.1 #2 Tue Jan 20 15:33:43 PST 2004 i686 GNU/Linux
> 
> locate did not find Xlib.h

You'll need to install xlibs-dev.

Pearu


From pearu at scipy.org  Fri Feb  6 03:28:28 2004
From: pearu at scipy.org (Pearu Peterson)
Date: Fri, 6 Feb 2004 02:28:28 -0600 (CST)
Subject: [SciPy-user] Re: need help with gcc error (Digest, Vol 5, Issue
 7)
In-Reply-To: <84d44854ec.854ec84d44@ucmerced.edu>
Message-ID: <Pine.LNX.4.44.0402060226140.3227-100000@scipy.org>



On Thu, 5 Feb 2004 ehaux at ucmerced.edu wrote:

> Also following up from Jan 27...
> 
> I get the following when using this script:
> 
> :/usr/lib/python2.3/site-packages/scipy$ python setup.py
> /usr/bin/python scipy_core/setup.py
> SciPy Core Version 0.2.1_alpha_40.1448
> usage: setup.py [global_opts] cmd1 [cmd1_opts] [cmd2 [cmd2_opts] ...]
>    or: setup.py --help [cmd1 cmd2 ...]
>    or: setup.py --help-commands
>    or: setup.py cmd --help
>  
> error: no commands supplied
  ^^^^^^^^^^^^^^^^^^^^^^^^^^^

Read:

  http://www.python.org/doc/current/inst/inst.html

Pearu


From christoph.ortner at comlab.ox.ac.uk  Fri Feb  6 05:19:35 2004
From: christoph.ortner at comlab.ox.ac.uk (Christoph Ortner)
Date: Fri, 06 Feb 2004 10:19:35 +0000
Subject: [SciPy-user] Sparse Matrices
Message-ID: <40236A37.8030307@comlab.ox.ac.uk>

Hi all,

Is there a chance that scipy will have support for sparse matrices
in the near future?
(e.g. pysparse -> http://people.web.psi.ch/geus/pyfemax/)

With the advantage, that a complete, (hopefully optimized) package
exists, optimization routines should be made able to use sparse
Hessians, etc.

Christoph


From bob at redivi.com  Fri Feb  6 17:50:44 2004
From: bob at redivi.com (Bob Ippolito)
Date: Fri, 6 Feb 2004 17:50:44 -0500
Subject: [SciPy-user] H E L P !!! PLEASE
In-Reply-To: <000f01c3e76c$c560de50$52806750@CASA>
References: <000f01c3e76c$c560de50$52806750@CASA>
Message-ID: <E58C737D-58F6-11D8-BF45-000A95686CD8@redivi.com>


On Jan 30, 2004, at 3:08 PM, Jos? A Mart?n H wrote:

> IDLE 1.0.2?????
>  >>> import scipy
> ?
> Traceback (most recent call last):
> ? File "<pyshell#0>", line 1, in -toplevel-
> ??? import scipy
> ? File "C:\Python23\Lib\site-packages\scipy\__init__.py", line 11, in 
> -toplevel-
> ??? from scipy_base import *
> ImportError: No module named scipy_base
> >>>
>  ?
> i have upgraded to python 2.3 and downloaded the last binaries for 
> scipy !!!.
>

I'm not a Windows user, but I have heard that this is a separate 
download (scipy_core maybe?).  In any case, you best bet might be to 
download the Enthought Python 2.3 distribution;  I believe that it 
includes both PIL and SciPy.

-bob


From pearu at scipy.org  Fri Feb  6 19:17:46 2004
From: pearu at scipy.org (Pearu Peterson)
Date: Fri, 6 Feb 2004 18:17:46 -0600 (CST)
Subject: [SciPy-user] H E L P !!! PLEASE
In-Reply-To: <000f01c3e76c$c560de50$52806750@CASA>
Message-ID: <Pine.LNX.4.44.0402061816270.26526-100000@scipy.org>



On Fri, 30 Jan 2004, Jos? A Mart?n H wrote:

> IDLE 1.0.2      
> >>> import scipy
> 
> Traceback (most recent call last):
>   File "<pyshell#0>", line 1, in -toplevel-
>     import scipy
>   File "C:\Python23\Lib\site-packages\scipy\__init__.py", line 11, in -toplevel-
>     from scipy_base import *
> ImportError: No module named scipy_base
> >>> 
> 
> i have upgraded to python 2.3 and downloaded the last binaries for scipy !!!.

You'll need to install scipy_core. I few day old snapshot is available at

  http://cens.ioc.ee/~pearu/scipy/

Pearu


From oliphant at ee.byu.edu  Fri Feb  6 07:49:23 2004
From: oliphant at ee.byu.edu (Travis E. Oliphant)
Date: Fri, 06 Feb 2004 05:49:23 -0700
Subject: [SciPy-user] Sparse Matrices
In-Reply-To: <40236A37.8030307@comlab.ox.ac.uk>
References: <40236A37.8030307@comlab.ox.ac.uk>
Message-ID: <40238D53.5010408@ee.byu.edu>

Christoph Ortner wrote:
> Hi all,
> 
> Is there a chance that scipy will have support for sparse matrices
> in the near future?
> (e.g. pysparse -> http://people.web.psi.ch/geus/pyfemax/)

It already does, but currently the automatic build on Windows is turned 
off (I'm not sure it builds on windows).

Sparse support on Linux works, however, though it's not well tested.



-Travis


From pearu at scipy.org  Sat Feb  7 13:24:45 2004
From: pearu at scipy.org (Pearu Peterson)
Date: Sat, 7 Feb 2004 12:24:45 -0600 (CST)
Subject: [SciPy-user] Sparse Matrices
In-Reply-To: <40238D53.5010408@ee.byu.edu>
Message-ID: <Pine.LNX.4.44.0402071213470.26526-100000@scipy.org>



On Fri, 6 Feb 2004, Travis E. Oliphant wrote:

> Christoph Ortner wrote:
> > Hi all,
> > 
> > Is there a chance that scipy will have support for sparse matrices
> > in the near future?
> > (e.g. pysparse -> http://people.web.psi.ch/geus/pyfemax/)
> 
> It already does, but currently the automatic build on Windows is turned 
> off (I'm not sure it builds on windows).
> 

scipy.sparse does not build on Windows
- with mingw because some of the sources in SuperLU2.0 use sys/times.h, 
  for instance, that is not available on non-posix system like Win32.
  This issue can be fixed by implementing sys/times.h features
  for windows.
- with cygwin because for some reasons _S and _C (if I remember correctly)
  in the definition of Dtype_t enum (in supermatrix.h) are defined
  by compiler and thus causing syntax error. This issue can be fixed by
  renaming Dtype_t items and replacing the corresponding occurrences 
  elsewhere with the new names.

Pearu


From christoph.ortner at comlab.ox.ac.uk  Sat Feb  7 18:55:35 2004
From: christoph.ortner at comlab.ox.ac.uk (Christoph Ortner)
Date: Sat, 07 Feb 2004 23:55:35 +0000
Subject: [SciPy-user] Sparse Matrices
In-Reply-To: <Pine.LNX.4.44.0402071213470.26526-100000@scipy.org>
References: <Pine.LNX.4.44.0402071213470.26526-100000@scipy.org>
Message-ID: <40257AF7.1010101@comlab.ox.ac.uk>



Pearu Peterson wrote:

>On Fri, 6 Feb 2004, Travis E. Oliphant wrote:
>
>  
>
>>Christoph Ortner wrote:
>>    
>>
>>>Hi all,
>>>
>>>Is there a chance that scipy will have support for sparse matrices
>>>in the near future?
>>>(e.g. pysparse -> http://people.web.psi.ch/geus/pyfemax/)
>>>      
>>>
>>It already does, but currently the automatic build on Windows is turned 
>>off (I'm not sure it builds on windows).
>>
>>    
>>
>
>scipy.sparse does not build on Windows
>- with mingw because some of the sources in SuperLU2.0 use sys/times.h, 
>  for instance, that is not available on non-posix system like Win32.
>  This issue can be fixed by implementing sys/times.h features
>  for windows.
>- with cygwin because for some reasons _S and _C (if I remember correctly)
>  in the definition of Dtype_t enum (in supermatrix.h) are defined
>  by compiler and thus causing syntax error. This issue can be fixed by
>  renaming Dtype_t items and replacing the corresponding occurrences 
>  elsewhere with the new names.
>
>Pearu
>  
>
Do you mean the pysparse package of  Roman Geus?
I built it both under cygwin and VC6.0 without any problems.
I didnt succed in mingw though, because I wasnt able to
create the python23.a.


From paxcalpt at sapo.pt  Sat Feb  7 20:02:19 2004
From: paxcalpt at sapo.pt (Ricardo Henriques)
Date: Sun, 8 Feb 2004 01:02:19 -0000
Subject: [SciPy-user] How to make an Gassian fit?
Message-ID: <01ae01c3eddf$35dc8440$0100a8c0@xcal>

Is it possible to make a Gaussian fit to histogram data using Scipy?

Tks everyone...

Paxcal


From pearu at scipy.org  Sun Feb  8 02:07:08 2004
From: pearu at scipy.org (Pearu Peterson)
Date: Sun, 8 Feb 2004 01:07:08 -0600 (CST)
Subject: [SciPy-user] Sparse Matrices
In-Reply-To: <40257AF7.1010101@comlab.ox.ac.uk>
Message-ID: <Pine.LNX.4.44.0402080102140.26526-100000@scipy.org>



On Sat, 7 Feb 2004, Christoph Ortner wrote:

> Pearu Peterson wrote:
> >scipy.sparse does not build on Windows
   ^^^^^

> Do you mean the pysparse package of  Roman Geus?

No. See ^^^ above.

> I built it both under cygwin and VC6.0 without any problems.
> I didnt succed in mingw though, because I wasnt able to
> create the python23.a.

Hint: when using scipy_distutils instead of distutils, libpython23.a
will be created for you. Also, google'ing gives number of references
how to create this file.

Pearu

> _______________________________________________
> SciPy-user mailing list
> SciPy-user at scipy.net
> http://www.scipy.net/mailman/listinfo/scipy-user
> 


From chris at fonnesbeck.org  Sun Feb  8 13:15:01 2004
From: chris at fonnesbeck.org (Christopher Fonnesbeck)
Date: Sun, 8 Feb 2004 13:15:01 -0500
Subject: [SciPy-user] extra_compile_args in weave
Message-ID: <B5C5B268-5A62-11D8-88F1-000A956FDAC0@fonnesbeck.org>

I am trying to write a GRASS GIS module (ANSI C) using weave.inline, so  
that I can interface my python simulation code with a GIS to get  
spatial information. As such, I need some GRASS header files. However,  
I am having trouble setting the GRASS include directory for weave. Here  
is a very simple example that simply initializes the GIS:

import os,weave

'Set GRASS environment variables'
os.putenv('GISBASE','/usr/local/grass57')
os.putenv('GISRC','$HOME/.grassrc57')
os.putenv('PATH','$PATH:$GISBASE/bin:$GISBASE/scripts')

def test(foo):

	code = """
	#include "gis.h"
	#include <stdio.h>
	#include <string.h>
	#include <math.h>
	
	G_gisinit(argv[0]);
	
	return foo;
	
	"""
	
	weave.inline(code,[foo],
		extra_compile_args=["-I/usr/local/grass57/include"])
		
if __name__=='__main__':

	test(3)


However, when I run this, I get:

Traceback (most recent call last):
   File "grasstest.py", line 27, in ?
     test(3)
   File "grasstest.py", line 23, in test
     extra_compile_args=["-I/usr/local/grass57/include"])
   File  
"/System/Library/Frameworks/Python.framework/Versions/2.3/lib/ 
python2.3/site-packages/weave/inline_tools.py", line 335, in inline
     auto_downcast = auto_downcast,
   File  
"/System/Library/Frameworks/Python.framework/Versions/2.3/lib/ 
python2.3/site-packages/weave/inline_tools.py", line 421, in  
compile_function
     type_converters = type_converters)
   File  
"/System/Library/Frameworks/Python.framework/Versions/2.3/lib/ 
python2.3/site-packages/weave/ext_tools.py", line 176, in __init__
     auto_downcast, type_converters)
   File  
"/System/Library/Frameworks/Python.framework/Versions/2.3/lib/ 
python2.3/site-packages/weave/ext_tools.py", line 391, in  
assign_variable_types
     errors[var] = ("The type and dimensionality specifications" +
TypeError: cannot concatenate 'str' and 'int' objects


It doesnt seem to like the extra_compile_args. How else am I to set  
this include directory?

Thanks,
Chris
--
Christopher J. Fonnesbeck ( c h r i s @ f o n n e s b e c k . o r g )
Georgia Cooperative Fish & Wildlife Research Unit, University of Georgia


From prabhu at aero.iitm.ernet.in  Sun Feb  8 13:36:53 2004
From: prabhu at aero.iitm.ernet.in (Prabhu Ramachandran)
Date: Mon, 9 Feb 2004 00:06:53 +0530
Subject: [SciPy-user] extra_compile_args in weave
In-Reply-To: <B5C5B268-5A62-11D8-88F1-000A956FDAC0@fonnesbeck.org>
References: <B5C5B268-5A62-11D8-88F1-000A956FDAC0@fonnesbeck.org>
Message-ID: <16422.33221.862765.834906@monster.linux.in>

>>>>> "CF" == Christopher Fonnesbeck <chris at fonnesbeck.org> writes:

    CF> I am trying to write a GRASS GIS module (ANSI C) using
    CF> weave.inline, so that I can interface my python simulation
    CF> code with a GIS to get spatial information. As such, I need
    CF> some GRASS header files. However, I am having trouble setting
    CF> the GRASS include directory for weave. Here is a very simple
    CF> example that simply initializes the GIS:

Try something like this:

def test(foo):
	code = """
	G_gisinit(argv[0]);
	
	return foo;	
	"""
	
	weave.inline(code, ['foo'], 
                     include_dirs=["/usr/local/grass57/include"],
                     headers=["gis.h", "stdio.h", "string.h", "math.h"])
		
All the options available can be seen here:

 http://www.scipy.org/site_content/weave/tutorial.html

in the Keyword options section.

I'm not sure your os.putenv statements are going to be of any use.
IIRC these changes do not propagate to your shell.

cheers,
prabhu


From chris at fonnesbeck.org  Sun Feb  8 13:50:18 2004
From: chris at fonnesbeck.org (Christopher Fonnesbeck)
Date: Sun, 8 Feb 2004 13:50:18 -0500
Subject: [SciPy-user] extra_compile_args in weave
In-Reply-To: <16422.33221.862765.834906@monster.linux.in>
References: <B5C5B268-5A62-11D8-88F1-000A956FDAC0@fonnesbeck.org>
	<16422.33221.862765.834906@monster.linux.in>
Message-ID: <A4041C22-5A67-11D8-88F1-000A956FDAC0@fonnesbeck.org>

On Feb 8, 2004, at 1:36 PM, Prabhu Ramachandran wrote:

>>>>>> "CF" == Christopher Fonnesbeck <chris at fonnesbeck.org> writes:
>
>     CF> I am trying to write a GRASS GIS module (ANSI C) using
>     CF> weave.inline, so that I can interface my python simulation
>     CF> code with a GIS to get spatial information. As such, I need
>     CF> some GRASS header files. However, I am having trouble setting
>     CF> the GRASS include directory for weave. Here is a very simple
>     CF> example that simply initializes the GIS:
>
> Try something like this:
>
> def test(foo):
> 	code = """
> 	G_gisinit(argv[0]);
> 	
> 	return foo;	
> 	"""
> 	
> 	weave.inline(code, ['foo'],
>                      include_dirs=["/usr/local/grass57/include"],
>                      headers=["gis.h", "stdio.h", "string.h",  
> "math.h"])
> 		

Thanks for the correction. However, I still get a similar error message:

Traceback (most recent call last):
   File "grasstest.py", line 19, in ?
     test(3)
   File "grasstest.py", line 15, in test
     headers=["gis.h", "stdio.h", "string.h", "math.h"])
   File  
"/System/Library/Frameworks/Python.framework/Versions/2.3/lib/ 
python2.3/site-packages/weave/inline_tools.py", line 335, in inline
     auto_downcast = auto_downcast,
   File  
"/System/Library/Frameworks/Python.framework/Versions/2.3/lib/ 
python2.3/site-packages/weave/inline_tools.py", line 421, in  
compile_function
     type_converters = type_converters)
   File  
"/System/Library/Frameworks/Python.framework/Versions/2.3/lib/ 
python2.3/site-packages/weave/ext_tools.py", line 176, in __init__
     auto_downcast, type_converters)
   File  
"/System/Library/Frameworks/Python.framework/Versions/2.3/lib/ 
python2.3/site-packages/weave/ext_tools.py", line 391, in  
assign_variable_types
     errors[var] = ("The type and dimensionality specifications" +
TypeError: cannot concatenate 'str' and 'int' objects

I  am using a recent CVS build of scipy (i.e. since weave was  
integrated into the scipy code base).

C.
--
Christopher J. Fonnesbeck ( c h r i s @ f o n n e s b e c k . o r g )
Georgia Cooperative Fish & Wildlife Research Unit, University of Georgia


From prabhu at aero.iitm.ernet.in  Mon Feb  9 12:45:39 2004
From: prabhu at aero.iitm.ernet.in (Prabhu Ramachandran)
Date: Mon, 9 Feb 2004 23:15:39 +0530
Subject: [SciPy-user] extra_compile_args in weave
In-Reply-To: <A4041C22-5A67-11D8-88F1-000A956FDAC0@fonnesbeck.org>
References: <B5C5B268-5A62-11D8-88F1-000A956FDAC0@fonnesbeck.org>
	<16422.33221.862765.834906@monster.linux.in>
	<A4041C22-5A67-11D8-88F1-000A956FDAC0@fonnesbeck.org>
Message-ID: <16423.51011.511913.836045@monster.linux.in>

>>>>> "CF" == Christopher Fonnesbeck <chris at fonnesbeck.org> writes:

    CF> On Feb 8, 2004, at 1:36 PM, Prabhu Ramachandran wrote:

[...]
    >> weave.inline(code, ['foo'],
                           ^^^^^

Note the quotes!  Your original code did not have the quotes and just
had [foo] which is incorrect.  I suspect (read: guess) that is your
problem.

prabhu


From jeremy.gore at yale.edu  Mon Feb  9 13:53:04 2004
From: jeremy.gore at yale.edu (Jeremy Gore)
Date: Mon, 9 Feb 2004 13:53:04 -0500
Subject: [SciPy-user] scipy OS X compiling on a G3
Message-ID: <31850D8C-5B31-11D8-8720-000A95843714@yale.edu>

I've tried all sorts of combinations of flags and so forth to get scipy  
to compile on OS X 10.3.2, but it still won't work for me.  With  
CFLAGS="-lcc_dynamic -framework vecLib" (and I've also tried with  
-faltivec and -framework Accelerate) set and using sudo python setup.py  
install (and have also tried with the build build_ext first) for the  
CVS version, I get error messages along the lines of:

In the build.py phase:
package init file  
'/Users/jmg75/Programming/scipy/scipy_core/scipy_distutils/tests/ 
__init__.py' not found (or not a regular file)
package init file  
'/Users/jmg75/Programming/scipy/scipy_core/scipy_base/tests/ 
__init__.py' not found (or not a regular file)

In the install_data phase:
fftw_info:
   NOT AVAILABLE
dfftw_info:
   NOT AVAILABLE
atlas_info:
scipy_distutils.system_info.atlas_info
   NOT AVAILABLE
blas_info:
   NOT AVAILABLE

Clearly it is not finding the vecLib BLAS libraries.  Whether this is  
because I'm on a G3 or some other problem I don't know; I was under the  
impression that although not optimized for altivec the vecLib libraries  
still processed for G3s.  Any suggestions?

My system:

Reading specs from  
/usr/local/lib/gcc/powerpc-apple-darwin7.2.0/3.4.0/specs
Configured with: ../gcc/configure --enable-threads=posix  
--enable-languages=f77
Thread model: posix
gcc version 3.4.0 20040107 (experimental)

Reading specs from /usr/libexec/gcc/darwin/ppc/3.3/specs
Thread model: posix
gcc version 3.3 20030304 (Apple Computer, Inc. build 1495)

on the run_all.py test for f2py:
Traceback (most recent call last):
   File  
"/Users/jmg75/Downloads/Expanded/F2PY-2.37.233-1545/tests/f90/ 
return_integer.py", line 136, in ?
     test_functions = build(f2py_opts)
   File  
"/Users/jmg75/Downloads/Expanded/F2PY-2.37.233-1545/tests/f90/ 
return_integer.py", line 16, in build
     assert not f2py2e.compile('''\
AssertionError
TEST FAILURE (status=1)

--
Jeremy Gore
1515 State St
New Haven, CT 06511
203-530-9157


From pearu at scipy.org  Mon Feb  9 14:37:03 2004
From: pearu at scipy.org (Pearu Peterson)
Date: Mon, 9 Feb 2004 13:37:03 -0600 (CST)
Subject: [SciPy-user] scipy OS X compiling on a G3
In-Reply-To: <31850D8C-5B31-11D8-8720-000A95843714@yale.edu>
Message-ID: <Pine.LNX.4.44.0402091332370.12125-100000@scipy.org>



On Mon, 9 Feb 2004, Jeremy Gore wrote:

> I've tried all sorts of combinations of flags and so forth to get scipy  
> to compile on OS X 10.3.2, but it still won't work for me.  With  
> CFLAGS="-lcc_dynamic -framework vecLib" (and I've also tried with  
> -faltivec and -framework Accelerate) set and using sudo python setup.py  
> install (and have also tried with the build build_ext first) for the  
> CVS version, I get error messages along the lines of:
> 
> In the build.py phase:
> package init file  
> '/Users/jmg75/Programming/scipy/scipy_core/scipy_distutils/tests/ 
> __init__.py' not found (or not a regular file)
> package init file  
> '/Users/jmg75/Programming/scipy/scipy_core/scipy_base/tests/ 
> __init__.py' not found (or not a regular file)
> 
> In the install_data phase:
> fftw_info:
>    NOT AVAILABLE
> dfftw_info:
>    NOT AVAILABLE
> atlas_info:
> scipy_distutils.system_info.atlas_info
>    NOT AVAILABLE
> blas_info:
>    NOT AVAILABLE
> 
> Clearly it is not finding the vecLib BLAS libraries.  Whether this is  
> because I'm on a G3 or some other problem I don't know; I was under the  
> impression that although not optimized for altivec the vecLib libraries  
> still processed for G3s.  Any suggestions?

Detecting vecLib BLAS libraries is not implemented in system_info.py yet.
Could you send me the information how one can detect vecLib OSX
and I'll add its support to system_info.py? E.g. which files it should 
look for, or which programs to run and what should be the expected output, 
the values of os.name and sys.platform, etc.

Pearu


From chris at fonnesbeck.org  Mon Feb  9 14:35:33 2004
From: chris at fonnesbeck.org (Christopher Fonnesbeck)
Date: Mon, 9 Feb 2004 14:35:33 -0500
Subject: [SciPy-user] scipy OS X compiling on a G3
In-Reply-To: <Pine.LNX.4.44.0402091332370.12125-100000@scipy.org>
References: <Pine.LNX.4.44.0402091332370.12125-100000@scipy.org>
Message-ID: <206F34D4-5B37-11D8-A63F-000A956FDAC0@fonnesbeck.org>


On Feb 9, 2004, at 2:37 PM, Pearu Peterson wrote:

>> Clearly it is not finding the vecLib BLAS libraries.  Whether this is
>> because I'm on a G3 or some other problem I don't know; I was under 
>> the
>> impression that although not optimized for altivec the vecLib 
>> libraries
>> still processed for G3s.  Any suggestions?
>
> Detecting vecLib BLAS libraries is not implemented in system_info.py 
> yet.
> Could you send me the information how one can detect vecLib OSX
> and I'll add its support to system_info.py? E.g. which files it should
> look for, or which programs to run and what should be the expected 
> output,
> the values of os.name and sys.platform, etc.

Adding the following compile flag does the trick for me:

-framework vecLib

C.,

--
Christopher J. Fonnesbeck ( c h r i s @ f o n n e s b e c k . o r g )
Georgia Cooperative Fish & Wildlife Research Unit, University of Georgia


From pearu at scipy.org  Mon Feb  9 14:45:20 2004
From: pearu at scipy.org (Pearu Peterson)
Date: Mon, 9 Feb 2004 13:45:20 -0600 (CST)
Subject: [SciPy-user] scipy OS X compiling on a G3
In-Reply-To: <206F34D4-5B37-11D8-A63F-000A956FDAC0@fonnesbeck.org>
Message-ID: <Pine.LNX.4.44.0402091342310.12125-100000@scipy.org>



On Mon, 9 Feb 2004, Christopher Fonnesbeck wrote:

> 
> On Feb 9, 2004, at 2:37 PM, Pearu Peterson wrote:
> 
> >> Clearly it is not finding the vecLib BLAS libraries.  Whether this is
> >> because I'm on a G3 or some other problem I don't know; I was under 
> >> the
> >> impression that although not optimized for altivec the vecLib 
> >> libraries
> >> still processed for G3s.  Any suggestions?
> >
> > Detecting vecLib BLAS libraries is not implemented in system_info.py 
> > yet.
> > Could you send me the information how one can detect vecLib OSX
> > and I'll add its support to system_info.py? E.g. which files it should
> > look for, or which programs to run and what should be the expected 
> > output,
> > the values of os.name and sys.platform, etc.
> 
> Adding the following compile flag does the trick for me:
> 
> -framework vecLib

Ok but will it work on all OSX systems?

Pearu


From chris at fonnesbeck.org  Mon Feb  9 14:54:21 2004
From: chris at fonnesbeck.org (Christopher Fonnesbeck)
Date: Mon, 9 Feb 2004 14:54:21 -0500
Subject: [SciPy-user] scipy OS X compiling on a G3
In-Reply-To: <Pine.LNX.4.44.0402091342310.12125-100000@scipy.org>
References: <Pine.LNX.4.44.0402091342310.12125-100000@scipy.org>
Message-ID: <C0C7AEA2-5B39-11D8-A63F-000A956FDAC0@fonnesbeck.org>


On Feb 9, 2004, at 2:45 PM, Pearu Peterson wrote:

>>
>>>> Clearly it is not finding the vecLib BLAS libraries.  Whether this 
>>>> is
>>>> because I'm on a G3 or some other problem I don't know; I was under
>>>> the
>>>> impression that although not optimized for altivec the vecLib
>>>> libraries
>>>> still processed for G3s.  Any suggestions?
>>>
>>> Detecting vecLib BLAS libraries is not implemented in system_info.py
>>> yet.
>>> Could you send me the information how one can detect vecLib OSX
>>> and I'll add its support to system_info.py? E.g. which files it 
>>> should
>>> look for, or which programs to run and what should be the expected
>>> output,
>>> the values of os.name and sys.platform, etc.
>>
>> Adding the following compile flag does the trick for me:
>>
>> -framework vecLib
>
> Ok but will it work on all OSX systems?

I believe so, as long as the developer tools are installed:

http://developer.apple.com/documentation/ReleaseNotes/MacOSX/vecLib.html

C.

--
Christopher J. Fonnesbeck ( c h r i s @ f o n n e s b e c k . o r g )
Georgia Cooperative Fish & Wildlife Research Unit, University of Georgia


From pearu at scipy.org  Mon Feb  9 15:07:30 2004
From: pearu at scipy.org (Pearu Peterson)
Date: Mon, 9 Feb 2004 14:07:30 -0600 (CST)
Subject: [SciPy-user] scipy OS X compiling on a G3
In-Reply-To: <C0C7AEA2-5B39-11D8-A63F-000A956FDAC0@fonnesbeck.org>
Message-ID: <Pine.LNX.4.44.0402091403120.12125-100000@scipy.org>



On Mon, 9 Feb 2004, Christopher Fonnesbeck wrote:

> >> Adding the following compile flag does the trick for me:
> >>
> >> -framework vecLib
> >
> > Ok but will it work on all OSX systems?
> 
> I believe so, as long as the developer tools are installed:
> 
> http://developer.apple.com/documentation/ReleaseNotes/MacOSX/vecLib.html

So, could you confirm that the following approach is correct:

if os.path.exists('/System/Library/Frameworks/vecLib.framework/'):
  # use the following compiler flags:
  #  -faltivec -framework vecLib
  # and no additional libraries need to be specified when linking.

?

Thanks,
Pearu


From chris at fonnesbeck.org  Mon Feb  9 15:23:25 2004
From: chris at fonnesbeck.org (Christopher Fonnesbeck)
Date: Mon, 9 Feb 2004 15:23:25 -0500
Subject: [SciPy-user] scipy OS X compiling on a G3
In-Reply-To: <Pine.LNX.4.44.0402091403120.12125-100000@scipy.org>
References: <Pine.LNX.4.44.0402091403120.12125-100000@scipy.org>
Message-ID: <D0A4074D-5B3D-11D8-A63F-000A956FDAC0@fonnesbeck.org>


On Feb 9, 2004, at 3:07 PM, Pearu Peterson wrote:

> So, could you confirm that the following approach is correct:
>
> if os.path.exists('/System/Library/Frameworks/vecLib.framework/'):
>   # use the following compiler flags:
>   #  -faltivec -framework vecLib
>   # and no additional libraries need to be specified when linking.

I see no reason why that shouldn't work ...

C.
--
Christopher J. Fonnesbeck ( c h r i s @ f o n n e s b e c k . o r g )
Georgia Cooperative Fish & Wildlife Research Unit, University of Georgia


From bob at redivi.com  Mon Feb  9 20:07:28 2004
From: bob at redivi.com (Bob Ippolito)
Date: Mon, 9 Feb 2004 20:07:28 -0500
Subject: [SciPy-user] scipy OS X compiling on a G3
In-Reply-To: <D0A4074D-5B3D-11D8-A63F-000A956FDAC0@fonnesbeck.org>
References: <Pine.LNX.4.44.0402091403120.12125-100000@scipy.org>
	<D0A4074D-5B3D-11D8-A63F-000A956FDAC0@fonnesbeck.org>
Message-ID: <7F189200-5B65-11D8-BF45-000A95686CD8@redivi.com>


On Feb 9, 2004, at 3:23 PM, Christopher Fonnesbeck wrote:

>
> On Feb 9, 2004, at 3:07 PM, Pearu Peterson wrote:
>
>> So, could you confirm that the following approach is correct:
>>
>> if os.path.exists('/System/Library/Frameworks/vecLib.framework/'):
>>   # use the following compiler flags:
>>   #  -faltivec -framework vecLib
>>   # and no additional libraries need to be specified when linking.
>
> I see no reason why that shouldn't work ...

If you don't have developer tools installed, you probably don't have a 
compiler anyway.

You should link to Accelerate.framework if it exists, and  vecLib 
otherwise.

-bob
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 2357 bytes
Desc: not available
URL: <http://mail.scipy.org/pipermail/scipy-user/attachments/20040209/f1942768/attachment.bin>

From chris at fonnesbeck.org  Mon Feb  9 23:29:52 2004
From: chris at fonnesbeck.org (Christopher Fonnesbeck)
Date: Mon, 9 Feb 2004 23:29:52 -0500
Subject: [SciPy-user] forcing linking in weave
Message-ID: <C52059B6-5B81-11D8-A63F-000A956FDAC0@fonnesbeck.org>

I need to link against a particular set of libraries for a weave.inline 
extension I am making. However, on compile, I get messages saying that 
linking is not done. How can I force linking with weave? I have 
specified include_dirs, library_dirs, libraries and headers, but to no 
avail.

Thanks for any advice,
C.

--
Christopher J. Fonnesbeck ( c h r i s @ f o n n e s b e c k . o r g )
Georgia Cooperative Fish & Wildlife Research Unit, University of Georgia


From scipy at zunzun.com  Tue Feb 10 05:08:35 2004
From: scipy at zunzun.com (James R. Phillips)
Date: Tue, 10 Feb 2004 05:08:35 -500
Subject: [SciPy-user] forcing linking in weave
References: <C52059B6-5B81-11D8-A63F-000A956FDAC0@fonnesbeck.org>
Message-ID: <4028ada3934f13.47178150@mercury.sabren.com>

> I need to link against a particular set of libraries for a weave.inline
> extension I am making.  How can I force linking with weave?

The version of the tutorial distributed with weave appears to be
more up-to-date than the online version.  It is located in the
Lib/site-packages/weave/doc directory of your Python install.

Look for the "Distutils keywords" section, where it refers
to lists of strings as in

          option = ['string1', 'string2']

For example, I use

          extra_compile_args = ['-O3']


>From the *offline* tutorial:

  inline() also accepts a number of distutils keywords for 
  controlling how the code is compiled.

    library_dirs
        [string] list of directories to search for C/C++ 
        libraries at link time 

    libraries
        [string] list of library names (not filenames or 
        paths) to link against 

    runtime_library_dirs
        [string] list of directories to search for C/C++
        libraries at run time (for shared extensions, 
        this is when the extension is loaded) 

    extra_objects
        [string] list of extra files to link with (eg.
        object files not implied by 'sources', static
        library that must be explicitly specified, binary
        resource files, etc.) 

    extra_compile_args
        [string] any extra platform- and compiler-specific
        information to use when compiling the source files
        in 'sources'. For platforms and compilers where 
        "command line" makes sense, this is typically a list
        of command-line arguments, but for other platforms
        it could be anything. 

    extra_link_args
        [string] any extra platform- and compiler-specific
        information to use when linking object files 
        together to create the extension (or to create a 
        new static Python interpreter). Similar 
        interpretation as for 'extra_compile_args'. 



     James Phillips
     http://zunzun.com


From pearu at scipy.org  Tue Feb 10 08:12:11 2004
From: pearu at scipy.org (Pearu Peterson)
Date: Tue, 10 Feb 2004 07:12:11 -0600 (CST)
Subject: [SciPy-user] scipy OS X compiling on a G3
In-Reply-To: <7F189200-5B65-11D8-BF45-000A95686CD8@redivi.com>
Message-ID: <Pine.LNX.4.44.0402100706510.12125-100000@scipy.org>



On Mon, 9 Feb 2004, Bob Ippolito wrote:

> 
> On Feb 9, 2004, at 3:23 PM, Christopher Fonnesbeck wrote:
> 
> >
> > On Feb 9, 2004, at 3:07 PM, Pearu Peterson wrote:
> >
> >> So, could you confirm that the following approach is correct:
> >>
> >> if os.path.exists('/System/Library/Frameworks/vecLib.framework/'):
> >>   # use the following compiler flags:
> >>   #  -faltivec -framework vecLib
> >>   # and no additional libraries need to be specified when linking.
> >
> > I see no reason why that shouldn't work ...
> 
> If you don't have developer tools installed, you probably don't have a 
> compiler anyway.
> 
> You should link to Accelerate.framework if it exists, and  vecLib 
> otherwise.

As you may have noticed then I am ignorant what comes to OS X. So, how 
your last remark applies to my suggestion? Is the following correct:

if os.path.exists('/System/Library/Frameworks/Accelerate.framework/'):
   # use the following compiler flags:
   #  -faltivec -framework Accelerate
   # and no additional libraries need to be specified when linking.
elif os.path.exists('/System/Library/Frameworks/vecLib.framework/'):
   # use the following compiler flags:
   #  -faltivec -framework vecLib
   # and no additional libraries need to be specified when linking.
?

Thanks,
Pearu


From bob at redivi.com  Tue Feb 10 08:26:24 2004
From: bob at redivi.com (Bob Ippolito)
Date: Tue, 10 Feb 2004 08:26:24 -0500
Subject: [SciPy-user] scipy OS X compiling on a G3
In-Reply-To: <Pine.LNX.4.44.0402100706510.12125-100000@scipy.org>
References: <Pine.LNX.4.44.0402100706510.12125-100000@scipy.org>
Message-ID: <B93E60B3-5BCC-11D8-BF45-000A95686CD8@redivi.com>


On Feb 10, 2004, at 8:12 AM, Pearu Peterson wrote:

>
>
> On Mon, 9 Feb 2004, Bob Ippolito wrote:
>
>>
>> On Feb 9, 2004, at 3:23 PM, Christopher Fonnesbeck wrote:
>>
>>>
>>> On Feb 9, 2004, at 3:07 PM, Pearu Peterson wrote:
>>>
>>>> So, could you confirm that the following approach is correct:
>>>>
>>>> if os.path.exists('/System/Library/Frameworks/vecLib.framework/'):
>>>>   # use the following compiler flags:
>>>>   #  -faltivec -framework vecLib
>>>>   # and no additional libraries need to be specified when linking.
>>>
>>> I see no reason why that shouldn't work ...
>>
>> If you don't have developer tools installed, you probably don't have a
>> compiler anyway.
>>
>> You should link to Accelerate.framework if it exists, and  vecLib
>> otherwise.
>
> As you may have noticed then I am ignorant what comes to OS X. So, how
> your last remark applies to my suggestion? Is the following correct:
>
> if os.path.exists('/System/Library/Frameworks/Accelerate.framework/'):
>    # use the following compiler flags:
>    #  -faltivec -framework Accelerate
>    # and no additional libraries need to be specified when linking.
> elif os.path.exists('/System/Library/Frameworks/vecLib.framework/'):
>    # use the following compiler flags:
>    #  -faltivec -framework vecLib
>    # and no additional libraries need to be specified when linking.
> ?

That is correct as far as I know.

-bob
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 2357 bytes
Desc: not available
URL: <http://mail.scipy.org/pipermail/scipy-user/attachments/20040210/b1d5122f/attachment.bin>

From pearu at scipy.org  Tue Feb 10 11:46:45 2004
From: pearu at scipy.org (Pearu Peterson)
Date: Tue, 10 Feb 2004 10:46:45 -0600 (CST)
Subject: [SciPy-user] "Building Scipy on Windows" notes available
Message-ID: <Pine.LNX.4.44.0402101035010.12125-100000@scipy.org>


Hi all,

Detailed "from scratch" notes about "Building Scipy on Windows" are 
available at

  http://cens.ioc.ee/~pearu/scipy/BUILD_WIN32.html

It also contains a section on building Win32 port of Atlas libraries 
that seems to be a major show-stopper for Windows people, and indeed, I found 
it a rather sensitive subject too. Hopefully these notes will help
Windows people to overcome the threshold of building Scipy themself and 
may enjoy most recent Scipy features from CVS.

Best regards,
	Pearu


From chris at fonnesbeck.org  Tue Feb 10 14:35:41 2004
From: chris at fonnesbeck.org (Christopher Fonnesbeck)
Date: Tue, 10 Feb 2004 14:35:41 -0500
Subject: [SciPy-user] zero-size graphic files from scipy.gplt
Message-ID: <4FE357AF-5C00-11D8-A63F-000A956FDAC0@fonnesbeck.org>

I am using scipy.gplt to generate plots of simulation time series. 
Every so often, however, plots that are saved to file using output() 
are of zero size, even though the plot looks fine on screen. Has anyone 
run into this problem before? There are no error messages, so this is a 
tough one to debug.

scipy cvs build on OSX 10.3 with python 2.3.3

Thanks,
Chris
--
Christopher J. Fonnesbeck ( c h r i s @ f o n n e s b e c k . o r g )
Georgia Cooperative Fish & Wildlife Research Unit, University of Georgia


From paxcalpt at sapo.pt  Wed Feb 11 14:50:45 2004
From: paxcalpt at sapo.pt (Ricardo Henriques)
Date: Wed, 11 Feb 2004 19:50:45 -0000
Subject: [SciPy-user] Where is my error when using scipy.optimize.leastsq
Message-ID: <001101c3f0d8$5effcb10$3eb6c151@xcal>

Hi...

Does anyone know how can I get the error (R**2) when using the
scipy.optimize.leastsq fitting.

Tks everyone...

Paxcal


From nwagner at mecha.uni-stuttgart.de  Fri Feb 13 07:30:36 2004
From: nwagner at mecha.uni-stuttgart.de (Nils Wagner)
Date: Fri, 13 Feb 2004 13:30:36 +0100
Subject: [SciPy-user] det(A-\lambda B) is identically zero
Message-ID: <402CC36C.89C2480F@mecha.uni-stuttgart.de>

Hi all,

Please find attached a small example for a singular pencil
A-\lambda B. det(A-\lambda B) is identically zero.
However linalg.eig(A,B) computes certain eigenvalues.
although \lambda can be chosen arbitrarely.

There should be at leat a warning in such a case.

Any comment or suggestion ?

Nils

from scipy import *
A = array((
[0,7,-1,1],
[4,2,3,1],
[2,0,0,2],
[6,9,2,4]))
B = array((
[3,4,1,2],
[-1,3,5,-6],
[1,0,0,1],
[15,7,6,9]))
print 'A'
print A
print 'B'
print B
#
# Eigenvalue problem
#
w,vr = linalg.eig(A,B)
for i in w:
 print i
#
# det(A-\lambda B) is identically zero
#
for l in arange(0,10.,0.1):
  d = linalg.det(A-l*B)
  print l,d


From pearu at scipy.org  Fri Feb 13 08:56:15 2004
From: pearu at scipy.org (Pearu Peterson)
Date: Fri, 13 Feb 2004 07:56:15 -0600 (CST)
Subject: [SciPy-user] det(A-\lambda B) is identically zero
In-Reply-To: <402CC36C.89C2480F@mecha.uni-stuttgart.de>
Message-ID: <Pine.LNX.4.44.0402130744210.5551-100000@scipy.org>



On Fri, 13 Feb 2004, Nils Wagner wrote:

> Please find attached a small example for a singular pencil
> A-\lambda B. det(A-\lambda B) is identically zero.
> However linalg.eig(A,B) computes certain eigenvalues.

And one of them is inf. That should make an user careful
when manipulating with (A,B) eigendata.

> although \lambda can be chosen arbitrarely.

Well, if lambda is small then det is close to small number indeed.
However note that

>>> linalg.det(A-1e5*B)
-966.29150970670446

for instance.

> There should be at leat a warning in such a case.

I think it is a computer numerics issue rather than of scipy.
What would be the criterion when such a message should be displayed?

Pearu


From h.jansen at fel.tno.nl  Fri Feb 13 10:08:27 2004
From: h.jansen at fel.tno.nl (H Jansen)
Date: Fri, 13 Feb 2004 16:08:27 +0100
Subject: [SciPy-user] ... object has no attribute 'zgelss'
Message-ID: <402CE86B.8050605@fel.tno.nl>

Many Lapack routines can't be found when I do:

import scipy
scipy.test(level=1)

For instance:

======================================================================
ERROR: check_random_complex_exact (test_basic.test_lstsq)
----------------------------------------------------------------------
Traceback (most recent call last):
   File 
"/opt/inst/python/2.3/lib/python2.3/site-packages/scipy/linalg/tests/test_basic.py", 
line 346, in check_random_complex_exact
     x = lstsq(a,b)[0]
   File 
"/opt/inst/python/2.3/lib/python2.3/site-packages/scipy/linalg/basic.py", 
line 281, in lstsq
     gelss, = get_lapack_funcs(('gelss',),(a1,b1))
   File 
"/opt/inst/python/2.3/lib/python2.3/site-packages/scipy/linalg/lapack.py", 
line 55, in get_lapack_funcs
     func = getattr(m2,func_name)
AttributeError: 'module' object has no attribute 'zgelss'

... << more of the same, all related to lapack subroutines >>

When I do "nm liblapack.a | grep zgelss" code for "zgelss_"; likewise 
with all other routines. I'm using ATLAS-3.6.0 and my liblapack is 
complete (appr. 6.8 Mb, routines added from CLAPACK). Apparently,
there's an interface mismatch between ATLAS and scipy 
(SciPy-0.2.0_alpha_200.4161).

The scipy source builts with the command

	python setup.py build build_ext `pkg-config atlas --cflags --libs`

resulting the linalg shared modules to be linked as

	-L/opt/src/math/numerical/matrix/atlas/current/lib/Linux_P4SSE2 
-llapack -lf77blas -lcblas -latlas -lg2c


Some help will be highly appreated! Thanks.

-- Henk


-- 
------------------------------------------------------------------------------
The disclaimer that applies to e-mail from
TNO Physics and Electronics Laboratory
can be found on: http://www.tno.nl/disclaimer/email.html
------------------------------------------------------------------------------


From nwagner at mecha.uni-stuttgart.de  Fri Feb 13 10:16:58 2004
From: nwagner at mecha.uni-stuttgart.de (Nils Wagner)
Date: Fri, 13 Feb 2004 16:16:58 +0100
Subject: [SciPy-user] Problems with latest cvs
Message-ID: <402CEA6A.3C229E0F@mecha.uni-stuttgart.de>

Hi all,

I have some trouble with latest cvs


Traceback (most recent call last):
  File "qr.py", line 79, in ?
    eval=linalg.eigvals(A)
  File "/usr/local/lib/python2.3/site-packages/scipy_base/ppimport.py",
line 243, in __getattr__
    module = self._ppimport_importer()
  File "/usr/local/lib/python2.3/site-packages/scipy_base/ppimport.py",
line 216, in _ppimport_importer
    module = __import__(name,None,None,['*'])
  File
"/usr/local/lib/python2.3/site-packages/scipy/linalg/__init__.py", line
8, in ?
    from basic import *
  File "/usr/local/lib/python2.3/site-packages/scipy/linalg/basic.py",
line 12, in ?
    from flinalg import get_flinalg_funcs
  File "/usr/local/lib/python2.3/site-packages/scipy/linalg/flinalg.py",
line 7, in ?
    from scipy_distutils.misc_util import PostponedException
  File
"/usr/local/lib/python2.3/site-packages/scipy_distutils/__init__.py",
line 14, in ?
    assert not sys.modules.has_key('distutils.ccompiler'),\
AssertionError: distutils has been imported before scipy_distutils
>>> scipy.__version__
'0.2.1_253.3711'
>>>

Nils


From pearu at scipy.org  Fri Feb 13 10:23:30 2004
From: pearu at scipy.org (Pearu Peterson)
Date: Fri, 13 Feb 2004 09:23:30 -0600 (CST)
Subject: [SciPy-user] ... object has no attribute 'zgelss'
In-Reply-To: <402CE86B.8050605@fel.tno.nl>
Message-ID: <Pine.LNX.4.44.0402130916360.5551-100000@scipy.org>



On Fri, 13 Feb 2004, H Jansen wrote:

> Many Lapack routines can't be found when I do:
> 
> import scipy
> scipy.test(level=1)
> 
> For instance:
<snip> 
> ... << more of the same, all related to lapack subroutines >>
> 
> When I do "nm liblapack.a | grep zgelss" code for "zgelss_";

??? So what is the result then? Are zgelss_ and others defined or not?

> likewise 
> with all other routines. I'm using ATLAS-3.6.0 and my liblapack is 
> complete (appr. 6.8 Mb, routines added from CLAPACK).

CLAPACK??? Unless CLAPACK provides symbol names with '_' suffix, you 
should complete ATLAS liblapack.a with Fortran LAPACK.

> Apparently,
> there's an interface mismatch between ATLAS and scipy 
> (SciPy-0.2.0_alpha_200.4161).
> 
> The scipy source builts with the command
> 
> 	python setup.py build build_ext `pkg-config atlas --cflags --libs`
> 
> resulting the linalg shared modules to be linked as
> 
> 	-L/opt/src/math/numerical/matrix/atlas/current/lib/Linux_P4SSE2 
> -llapack -lf77blas -lcblas -latlas -lg2c

I suspect that scipy setup finds different atlas/lapack from you system 
than that you specify with `pkd-config`. Try building scipy as follows

  rm -rf build
  export ATLAS=/opt/src/math/numerical/matrix/atlas/current/lib/Linux_P4SSE2
  python setup.py build

HTH,
Pearu


From pearu at scipy.org  Fri Feb 13 10:30:17 2004
From: pearu at scipy.org (Pearu Peterson)
Date: Fri, 13 Feb 2004 09:30:17 -0600 (CST)
Subject: [SciPy-user] Problems with latest cvs
In-Reply-To: <402CEA6A.3C229E0F@mecha.uni-stuttgart.de>
Message-ID: <Pine.LNX.4.44.0402130928200.5551-100000@scipy.org>



On Fri, 13 Feb 2004, Nils Wagner wrote:

> I have some trouble with latest cvs
> 
>     assert not sys.modules.has_key('distutils.ccompiler'),\
> AssertionError: distutils has been imported before scipy_distutils

So, have you imported distutils before scipy_distutils or scipy?
The fix depends on the answer..

Thanks,
Pearu


From nwagner at mecha.uni-stuttgart.de  Fri Feb 13 10:28:03 2004
From: nwagner at mecha.uni-stuttgart.de (Nils Wagner)
Date: Fri, 13 Feb 2004 16:28:03 +0100
Subject: [SciPy-user] Problems with latest cvs
References: <Pine.LNX.4.44.0402130928200.5551-100000@scipy.org>
Message-ID: <402CED03.864AD1B2@mecha.uni-stuttgart.de>

Pearu Peterson schrieb:
> 
> On Fri, 13 Feb 2004, Nils Wagner wrote:
> 
> > I have some trouble with latest cvs
> >
> >     assert not sys.modules.has_key('distutils.ccompiler'),\
> > AssertionError: distutils has been imported before scipy_distutils
> 
> So, have you imported distutils before scipy_distutils or scipy?
> The fix depends on the answer..
> 
> Thanks,
> Pearu
> 
> _________

Pearu,

just 

from scipy import *


Nils

______________________________________
> SciPy-user mailing list
> SciPy-user at scipy.net
> http://www.scipy.net/mailman/listinfo/scipy-user


From pearu at scipy.org  Fri Feb 13 10:46:58 2004
From: pearu at scipy.org (Pearu Peterson)
Date: Fri, 13 Feb 2004 09:46:58 -0600 (CST)
Subject: [SciPy-user] Problems with latest cvs
In-Reply-To: <402CED03.864AD1B2@mecha.uni-stuttgart.de>
Message-ID: <Pine.LNX.4.44.0402130945210.5551-100000@scipy.org>



On Fri, 13 Feb 2004, Nils Wagner wrote:

> > So, have you imported distutils before scipy_distutils or scipy?
> > The fix depends on the answer..
> 
> just 
> 
> from scipy import *

Strange, the site.py or something has imported distutils for you.
It does not happend here. Anyway, a fix is in CVS. You'll only need to 
reinstall scipy_core.

Pearu


From nwagner at mecha.uni-stuttgart.de  Fri Feb 13 10:51:46 2004
From: nwagner at mecha.uni-stuttgart.de (Nils Wagner)
Date: Fri, 13 Feb 2004 16:51:46 +0100
Subject: [SciPy-user] Problems with latest cvs
References: <Pine.LNX.4.44.0402130945210.5551-100000@scipy.org>
Message-ID: <402CF292.850BA75E@mecha.uni-stuttgart.de>

Pearu Peterson schrieb:
> 
> On Fri, 13 Feb 2004, Nils Wagner wrote:
> 
> > > So, have you imported distutils before scipy_distutils or scipy?
> > > The fix depends on the answer..
> >
> > just
> >
> > from scipy import *
> 
> Strange, the site.py or something has imported distutils for you.
> It does not happend here. Anyway, a fix is in CVS. You'll only need to
> reinstall scipy_core.
> 
> Pearu
> 
> _______________________________________________
> SciPy-user mailing list
> SciPy-user at scipy.net
> http://www.scipy.net/mailman/listinfo/scipy-user

Sorry, but the problem is still there.

Nils
Traceback (most recent call last):
  File "qr.py", line 89, in ?
    eval=linalg.eigvals(A)
  File "/usr/local/lib/python2.3/site-packages/scipy_base/ppimport.py",
line 243, in __getattr__
    module = self._ppimport_importer()
  File "/usr/local/lib/python2.3/site-packages/scipy_base/ppimport.py",
line 216, in _ppimport_importer
    module = __import__(name,None,None,['*'])
  File
"/usr/local/lib/python2.3/site-packages/scipy/linalg/__init__.py", line
8, in ?
    from basic import *
  File "/usr/local/lib/python2.3/site-packages/scipy/linalg/basic.py",
line 12, in ?
    from flinalg import get_flinalg_funcs
  File "/usr/local/lib/python2.3/site-packages/scipy/linalg/flinalg.py",
line 7, in ?
    from scipy_distutils.misc_util import PostponedException
  File
"/usr/local/lib/python2.3/site-packages/scipy_distutils/__init__.py",
line 14, in ?
    assert not sys.modules.has_key('distutils.ccompiler'),\
AssertionError: distutils has been imported before scipy_distutils


From pearu at scipy.org  Fri Feb 13 11:02:05 2004
From: pearu at scipy.org (Pearu Peterson)
Date: Fri, 13 Feb 2004 10:02:05 -0600 (CST)
Subject: [SciPy-user] Problems with latest cvs
In-Reply-To: <402CF292.850BA75E@mecha.uni-stuttgart.de>
Message-ID: <Pine.LNX.4.44.0402130959090.5551-100000@scipy.org>



On Fri, 13 Feb 2004, Nils Wagner wrote:

> Sorry, but the problem is still there.

Impossible..

>   File
> "/usr/local/lib/python2.3/site-packages/scipy_distutils/__init__.py",
> line 14, in ?
>     assert not sys.modules.has_key('distutils.ccompiler'),\
> AssertionError: distutils has been imported before scipy_distutils

In CVS version of __init__.py line #14 contains a comment line and the 
assert statement is on line #15.
So, you have failed to update scipy_core.

Pearu


From h.jansen at fel.tno.nl  Fri Feb 13 10:58:47 2004
From: h.jansen at fel.tno.nl (H Jansen)
Date: Fri, 13 Feb 2004 16:58:47 +0100
Subject: [SciPy-user] ... object has no attribute 'zgelss'
In-Reply-To: <Pine.LNX.4.44.0402130916360.5551-100000@scipy.org>
References: <Pine.LNX.4.44.0402130916360.5551-100000@scipy.org>
Message-ID: <1076687927.27613.16.camel@pc50002760.fel.tno.nl>

On Fri, 2004-02-13 at 16:23, Pearu Peterson wrote:
> On Fri, 13 Feb 2004, H Jansen wrote:
> 
> > Many Lapack routines can't be found when I do:
> > 
> > import scipy
> > scipy.test(level=1)
> > 
> > For instance:
> <snip> 
> > ... << more of the same, all related to lapack subroutines >>
> > 
> > When I do "nm liblapack.a | grep zgelss" code for "zgelss_";
> 
> ??? So what is the result then? Are zgelss_ and others defined or not?
> 
> > likewise 
> > with all other routines. I'm using ATLAS-3.6.0 and my liblapack is 
> > complete (appr. 6.8 Mb, routines added from CLAPACK).
> 
> CLAPACK??? Unless CLAPACK provides symbol names with '_' suffix, you 
> should complete ATLAS liblapack.a with Fortran LAPACK.

With Fortran Lapack everything now works fine!

Thanks! Henk.


> > Apparently,
> > there's an interface mismatch between ATLAS and scipy 
> > (SciPy-0.2.0_alpha_200.4161).
> > 
> > The scipy source builts with the command
> > 
> > 	python setup.py build build_ext `pkg-config atlas --cflags --libs`
> > 
> > resulting the linalg shared modules to be linked as
> > 
> > 	-L/opt/src/math/numerical/matrix/atlas/current/lib/Linux_P4SSE2 
> > -llapack -lf77blas -lcblas -latlas -lg2c
> 
> I suspect that scipy setup finds different atlas/lapack from you system 
> than that you specify with `pkd-config`. Try building scipy as follows
> 
>   rm -rf build
>   export ATLAS=/opt/src/math/numerical/matrix/atlas/current/lib/Linux_P4SSE2
>   python setup.py build
> 
> HTH,
> Pearu
> 
> _______________________________________________
> SciPy-user mailing list
> SciPy-user at scipy.net
> http://www.scipy.net/mailman/listinfo/scipy-user
-- 
Henk Jansen <h.jansen at fel.tno.nl>
TNO Physics and Electronics Laboratory


-- 
------------------------------------------------------------------------------
The disclaimer that applies to e-mail from
TNO Physics and Electronics Laboratory
can be found on: http://www.tno.nl/disclaimer/email.html
------------------------------------------------------------------------------


From j_r_fonseca at yahoo.co.uk  Sun Feb 15 14:16:18 2004
From: j_r_fonseca at yahoo.co.uk (=?iso-8859-1?Q?Jos=E9?= Fonseca)
Date: Sun, 15 Feb 2004 19:16:18 +0000 (UTC)
Subject: [SciPy-user] New Scipy Debian packages - Testers wanted!
Message-ID: <c0ogi1$la1$1@sea.gmane.org>

New Scipy & F2PY Debian packages are available at
<http://jrfonseca.dyndns.org/debian/>.

These no longer have Pentium 4 specific code; should compile perfectly
from sources; and no longer get any warning from lintian. Note that
chaco no longer is included though.

They should be almost ready for inclusion in the official Debian
repository, but before that I'd like them to be more widely tested. So
please test and let me know (privately) if they work ok with you!

Many thanks to Pearu Peterson for all his efforts in making the
packaging much easier.

Jose Fonseca


From arnd.baecker at web.de  Mon Feb 16 10:26:29 2004
From: arnd.baecker at web.de (Arnd Baecker)
Date: Mon, 16 Feb 2004 16:26:29 +0100 (CET)
Subject: [SciPy-user] scipy.xplt.ghelp missing.
Message-ID: <Pine.LNX.4.51.0402161625140.4164@ptpcp2.phy.tu-dresden.de>

Hi,

after a recent CVS download  scipy.xplt.ghelp
is missing. I tracked this down to
a removed "from helpmod import help as ghelp"
in gist.py (see below).

Of course I think it would be great if the xplt documentation
gets integrated into the docstrings so that
tools like pydoc can make use of it
(Also it would be one `help` type command less to `teach`
to the students).

For the commmands themselves this works fine, e.g.
  scipy.xplt.ghelp("plg")
vs.
  help("scipy.xplt.plg")
but for the keywords, e.g.
  scipy.xplt.ghelp("type")
I can't find any equivalent with the "normal" help.

Well, so did I just do a check out during some reorganization
of the help for scipy.xplt
or should the `from helpmod import help as ghelp`
be put back into gist.py ?

Many thanks,

Arnd


__version__ = "$Id: gist.py,v 1.15 2003/06/25 11:02:01 travo Exp $"

from gistC import *
from helpmod import help as ghelp
from pydoc import help


__version__ = "1.5.18"

from gistC import *
from pydoc import help


From scipy at zunzun.com  Mon Feb 16 15:13:00 2004
From: scipy at zunzun.com (James R. Phillips)
Date: Mon, 16 Feb 2004 15:13:00 -500
Subject: [SciPy-user] Night of the Living Wgnuplot.exe
Message-ID: <4031244c256b35.29068471@mercury.sabren.com>

I have a lot of undying wgnuplot.exe processes
on my XP box that I must manually kill, they
appear each time I use scipy.  Has anyone
else seen this?

     James Phillips
     http://zunzun.com


From jdhunter at ace.bsd.uchicago.edu  Tue Feb 17 09:19:21 2004
From: jdhunter at ace.bsd.uchicago.edu (John Hunter)
Date: Tue, 17 Feb 2004 08:19:21 -0600
Subject: [SciPy-user] ANN: matplotlib-0.50 - python plotting
Message-ID: <m2u11pss9i.fsf@mother.paradise.lost>


matplotlib is a python 2D plotting library which produces publication
quality figures using in a variety of hardcopy formats (PNG, JPG,
TIFF, PS) and interactive GUI environments (WX, GTK) across platforms.
matplotlib can be used in python scripts, interactively from the
python shell (ala matlab or mathematica), in web application servers
generating dynamic charts, or embedded in GTK or WX applications.

  http://matplotlib.sourceforge.net

What's new in matplotlib 0.50

Antigrain backend: Agg

  Adding to the growing list of image backends is Antigrain --
  http://antigrain.com. This is a backend written mostly in extension
  code and is the fastest of all the image backends. Agg supports
  freetype fonts, antialiased drawing, alpha blending, and much
  more. The windows installer contains everything you need except
  Numeric to use the agg backend out of the box; for other platforms
  see http://matplotlib.sourceforge.net/backends.html#Agg

Paint/libart backend

  David Moore wrote a backend for pypaint, a libart wrapper. libart is
  a high quality, cross platform image renderer that supports
  antialiased lines, freetype fonts, and other capabilities to soon be
  exploited. Thanks David! See
  http://matplotlib.sourceforge.net/backends.html#Paint for more
  information and install instructions

The Matplotlib FAQ

  Matplotlib now has a FAQ -- http://matplotlib.sourceforge.net/faq.html

Alpha channel attribute

  All the figure elements now have an alpha attribute to allow
  blending and translucency. Not all backends are currenly capable of
  supporting alpha - currently only Agg, but Paint should be able to
  support this soon - see the scatter screenshot for an example of
  alpha at work
  http://matplotlib.sourceforge.net/screenshots.html#scatter_demo2

Table class added

  John Gill has developed a very nice Table class and table function
  that plays well with bar charts and stacked bar charts. See example
  code and screenshot table_demo at
  http://matplotlib.sourceforge.net/screenshots.html#table_demo

New plot commands cla and clf

  Clear the current axes or figure. Useful in interactive plotting
  from a python shell

GD module on win32

  With much weeping and gnashing of teeth and help from half the
  people on this globe, built a gdmodule win32 installer. Special
  thanks to Stefan Kuzminski for putting up with my endless windows
  confusions. See the win32 quickstart at installing the GD backend -
  http://matplotlib.sourceforge.net/backends.html#GDWIN32

GD supports clipping and antialiased line drawing

  See instructions about upgrading gd and gdmodule at Installing the
  GD backend. The line object has a new 'antialiased' property, that
  if True, the backend will render the line antialiased if
  supported. Note antialiased drawing under GD is slow, so be sure to
  turn the property off set(lines, 'antialiased', False) if you
  experience performance problems. If you need performance and
  antialiasing, use the agg backend.

Wild and wonderful bar charts

  You can provide an optional argument bottom to the bar command to
  determine where the bottom of each bar is, default 0 for all. This
  enables stacked bar plots and candelstick plots --
  examples/bar_stacked.py. Thanks to David Moore and John Gill for
  suggestions and code.

Figure backend refactored

  The figure functionality was split into a backend independent
  component Figure and a backend dependent component
  FigureCanvasBase. This completes the transition to a totally
  abstract figure interface and improves the ability the switch
  backends and a figure to multiple backends. See API_CHANGES for
  information on migrating applications to the new API at
  http://matplotlib.sourceforge.net/API_CHANGES

Tons of bug fixes and optimizations detailed at
http://matplotlib.sourceforge.net/whats_new.html


From chris at fonnesbeck.org  Tue Feb 17 09:58:18 2004
From: chris at fonnesbeck.org (Christopher Fonnesbeck)
Date: Tue, 17 Feb 2004 09:58:18 -0500
Subject: [SciPy-user] fftpack fails on cvs build (OSX)
Message-ID: <B8657FB6-6159-11D8-86FB-000A956FDAC0@fonnesbeck.org>

After updating from cvs and building, I get the following error:

gcc options: "-fno-strict-aliasing -Wno-long-double -no-cpp-precomp  
-mno-fused-madd -fno-common -dynamic -DNDEBUG -g -O3 -Wall  
-Wstrict-prototypes -faltivec"
compile options: '-DSCIPY_FFTW_H -I/usr/local/include -Ibuild/src  
-I/System/Library/Frameworks/Python.framework/Versions/2.3/include/ 
python2.3 -c'
/usr/local/bin/g77  
build/temp.darwin-7.2.0-Power_Macintosh-2.3/build/src/Lib/fftpack/ 
_fftpackmodule.o  
build/temp.darwin-7.2.0-Power_Macintosh-2.3/Lib/fftpack/src/zfft.o  
build/temp.darwin-7.2.0-Power_Macintosh-2.3/Lib/fftpack/src/drfft.o  
build/temp.darwin-7.2.0-Power_Macintosh-2.3/Lib/fftpack/src/zrfft.o  
build/temp.darwin-7.2.0-Power_Macintosh-2.3/Lib/fftpack/src/zfftnd.o  
build/temp.darwin-7.2.0-Power_Macintosh-2.3/build/src/fortranobject.o  
-L/usr/local/lib -L/usr/local/lib/gcc/powerpc-apple-darwin7.0.0/3.4  
-Lbuild/temp.darwin-7.2.0-Power_Macintosh-2.3 -ldfftpack -lrfftw -lfftw  
-lg2c -o  
build/lib.darwin-7.2.0-Power_Macintosh-2.3/scipy/fftpack/_fftpack.so
/usr/bin/ld: Undefined symbols:
_PyArg_ParseTuple
_PyArg_ParseTupleAndKeywords
_PyCObject_AsVoidPtr
_PyCObject_Type
_PyComplex_Type
_PyDict_GetItemString
_PyDict_SetItemString
_PyErr_Clear
_PyErr_NewException
_PyErr_Occurred
_PyErr_Print
_PyErr_SetString
_PyImport_ImportModule
_PyInt_Type
_PyModule_GetDict
_PyNumber_Int
_PyObject_GetAttrString
_PySequence_Check
_PySequence_GetItem
_PyString_FromString
_PyString_Type
_PyType_IsSubtype
_PyType_Type
_Py_BuildValue
_Py_FatalError
_Py_InitModule4
__Py_NoneStruct
_PyDict_DelItemString
_PyDict_New
_PyErr_Format
_PyExc_AttributeError
_PyExc_RuntimeError
_PyExc_TypeError
_PyObject_Free
_PyString_ConcatAndDel
_Py_FindMethod
__PyObject_New
restFP
saveFP
_MAIN__
collect2: ld returned 1 exit status
error: Command "/usr/local/bin/g77  
build/temp.darwin-7.2.0-Power_Macintosh-2.3/build/src/Lib/fftpack/ 
_fftpackmodule.o  
build/temp.darwin-7.2.0-Power_Macintosh-2.3/Lib/fftpack/src/zfft.o  
build/temp.darwin-7.2.0-Power_Macintosh-2.3/Lib/fftpack/src/drfft.o  
build/temp.darwin-7.2.0-Power_Macintosh-2.3/Lib/fftpack/src/zrfft.o  
build/temp.darwin-7.2.0-Power_Macintosh-2.3/Lib/fftpack/src/zfftnd.o  
build/temp.darwin-7.2.0-Power_Macintosh-2.3/build/src/fortranobject.o  
-L/usr/local/lib -L/usr/local/lib/gcc/powerpc-apple-darwin7.0.0/3.4  
-Lbuild/temp.darwin-7.2.0-Power_Macintosh-2.3 -ldfftpack -lrfftw -lfftw  
-lg2c -o  
build/lib.darwin-7.2.0-Power_Macintosh-2.3/scipy/fftpack/_fftpack.so"  
failed with exit status 1

Not sure why -- worked fine before the update.

C.
--
Christopher J. Fonnesbeck ( c h r i s @ f o n n e s b e c k . o r g )
Georgia Cooperative Fish & Wildlife Research Unit, University of Georgia


From bob at redivi.com  Tue Feb 17 10:09:40 2004
From: bob at redivi.com (Bob Ippolito)
Date: Tue, 17 Feb 2004 10:09:40 -0500
Subject: [SciPy-user] fftpack fails on cvs build (OSX)
In-Reply-To: <B8657FB6-6159-11D8-86FB-000A956FDAC0@fonnesbeck.org>
References: <B8657FB6-6159-11D8-86FB-000A956FDAC0@fonnesbeck.org>
Message-ID: <4F62AE0C-615B-11D8-B0CA-000A95686CD8@redivi.com>


On Feb 17, 2004, at 9:58 AM, Christopher Fonnesbeck wrote:

> After updating from cvs and building, I get the following error:
>
> gcc options: "-fno-strict-aliasing -Wno-long-double -no-cpp-precomp  
> -mno-fused-madd -fno-common -dynamic -DNDEBUG -g -O3 -Wall  
> -Wstrict-prototypes -faltivec"
> compile options: '-DSCIPY_FFTW_H -I/usr/local/include -Ibuild/src  
> -I/System/Library/Frameworks/Python.framework/Versions/2.3/include/ 
> python2.3 -c'
> /usr/local/bin/g77  
> build/temp.darwin-7.2.0-Power_Macintosh-2.3/build/src/Lib/fftpack/ 
> _fftpackmodule.o  
> build/temp.darwin-7.2.0-Power_Macintosh-2.3/Lib/fftpack/src/zfft.o  
> build/temp.darwin-7.2.0-Power_Macintosh-2.3/Lib/fftpack/src/drfft.o  
> build/temp.darwin-7.2.0-Power_Macintosh-2.3/Lib/fftpack/src/zrfft.o  
> build/temp.darwin-7.2.0-Power_Macintosh-2.3/Lib/fftpack/src/zfftnd.o  
> build/temp.darwin-7.2.0-Power_Macintosh-2.3/build/src/fortranobject.o  
> -L/usr/local/lib -L/usr/local/lib/gcc/powerpc-apple-darwin7.0.0/3.4  
> -Lbuild/temp.darwin-7.2.0-Power_Macintosh-2.3 -ldfftpack -lrfftw  
> -lfftw -lg2c -o  
> build/lib.darwin-7.2.0-Power_Macintosh-2.3/scipy/fftpack/_fftpack.so
> /usr/bin/ld: Undefined symbols:
> _PyArg_ParseTuple
> ....

Extensions need to link to Python (-framework Python).

-bob


From pearu at scipy.org  Tue Feb 17 10:13:48 2004
From: pearu at scipy.org (Pearu Peterson)
Date: Tue, 17 Feb 2004 09:13:48 -0600 (CST)
Subject: [SciPy-user] fftpack fails on cvs build (OSX)
In-Reply-To: <B8657FB6-6159-11D8-86FB-000A956FDAC0@fonnesbeck.org>
Message-ID: <Pine.LNX.4.44.0402170909010.26316-100000@scipy.org>



On Tue, 17 Feb 2004, Christopher Fonnesbeck wrote:

> After updating from cvs and building, I get the following error:
> 
> gcc options: "-fno-strict-aliasing -Wno-long-double -no-cpp-precomp  
> -mno-fused-madd -fno-common -dynamic -DNDEBUG -g -O3 -Wall  
> -Wstrict-prototypes -faltivec"
> compile options: '-DSCIPY_FFTW_H -I/usr/local/include -Ibuild/src  
> -I/System/Library/Frameworks/Python.framework/Versions/2.3/include/ 
> python2.3 -c'
> /usr/local/bin/g77  
> build/temp.darwin-7.2.0-Power_Macintosh-2.3/build/src/Lib/fftpack/ 
> _fftpackmodule.o  
> build/temp.darwin-7.2.0-Power_Macintosh-2.3/Lib/fftpack/src/zfft.o  
> build/temp.darwin-7.2.0-Power_Macintosh-2.3/Lib/fftpack/src/drfft.o  
> build/temp.darwin-7.2.0-Power_Macintosh-2.3/Lib/fftpack/src/zrfft.o  
> build/temp.darwin-7.2.0-Power_Macintosh-2.3/Lib/fftpack/src/zfftnd.o  
> build/temp.darwin-7.2.0-Power_Macintosh-2.3/build/src/fortranobject.o  
> -L/usr/local/lib -L/usr/local/lib/gcc/powerpc-apple-darwin7.0.0/3.4  
> -Lbuild/temp.darwin-7.2.0-Power_Macintosh-2.3 -ldfftpack -lrfftw -lfftw  
> -lg2c -o  
> build/lib.darwin-7.2.0-Power_Macintosh-2.3/scipy/fftpack/_fftpack.so

Could you run this command with -shared option? Does it work?

Btw, previous gnu fcompiler implementation had a comment "Darwin g77 
cannot be used as a linker", is it really true and what could be the 
backround of this comment if this is true?

Pearu


From pearu at scipy.org  Tue Feb 17 10:26:11 2004
From: pearu at scipy.org (Pearu Peterson)
Date: Tue, 17 Feb 2004 09:26:11 -0600 (CST)
Subject: [SciPy-user] fftpack fails on cvs build (OSX)
In-Reply-To: <B8657FB6-6159-11D8-86FB-000A956FDAC0@fonnesbeck.org>
Message-ID: <Pine.LNX.4.44.0402170924580.26316-100000@scipy.org>



On Tue, 17 Feb 2004, Christopher Fonnesbeck wrote:

> After updating from cvs and building, I get the following error:
> 
> gcc options: "-fno-strict-aliasing -Wno-long-double -no-cpp-precomp  
> -mno-fused-madd -fno-common -dynamic -DNDEBUG -g -O3 -Wall  
> -Wstrict-prototypes -faltivec"
> compile options: '-DSCIPY_FFTW_H -I/usr/local/include -Ibuild/src  
> -I/System/Library/Frameworks/Python.framework/Versions/2.3/include/ 
> python2.3 -c'
> /usr/local/bin/g77  
> build/temp.darwin-7.2.0-Power_Macintosh-2.3/build/src/Lib/fftpack/ 
> _fftpackmodule.o  
> build/temp.darwin-7.2.0-Power_Macintosh-2.3/Lib/fftpack/src/zfft.o  
> build/temp.darwin-7.2.0-Power_Macintosh-2.3/Lib/fftpack/src/drfft.o  
> build/temp.darwin-7.2.0-Power_Macintosh-2.3/Lib/fftpack/src/zrfft.o  
> build/temp.darwin-7.2.0-Power_Macintosh-2.3/Lib/fftpack/src/zfftnd.o  
> build/temp.darwin-7.2.0-Power_Macintosh-2.3/build/src/fortranobject.o  
> -L/usr/local/lib -L/usr/local/lib/gcc/powerpc-apple-darwin7.0.0/3.4  
> -Lbuild/temp.darwin-7.2.0-Power_Macintosh-2.3 -ldfftpack -lrfftw -lfftw  
> -lg2c -o  
> build/lib.darwin-7.2.0-Power_Macintosh-2.3/scipy/fftpack/_fftpack.so
> /usr/bin/ld: Undefined symbols:
> _PyArg_ParseTuple

Could you update cvs and try again? I have added Bob's suggestion there.
Let me know if it works or not.

Thanks,
Pearu


From chris at fonnesbeck.org  Tue Feb 17 10:36:27 2004
From: chris at fonnesbeck.org (Christopher Fonnesbeck)
Date: Tue, 17 Feb 2004 10:36:27 -0500
Subject: [SciPy-user] fftpack fails on cvs build (OSX)
In-Reply-To: <Pine.LNX.4.44.0402170924580.26316-100000@scipy.org>
References: <Pine.LNX.4.44.0402170924580.26316-100000@scipy.org>
Message-ID: <0CB9A903-615F-11D8-86FB-000A956FDAC0@fonnesbeck.org>


On Feb 17, 2004, at 10:26 AM, Pearu Peterson wrote:

> Could you update cvs and try again? I have added Bob's suggestion 
> there.
> Let me know if it works or not.

You added:

opt.append('-framework','Python')

which gives an error, since append only takes 1 argument. Substituting:

opt.append("-framework Python")

gives the following error:

g77: Python: No such file or directory

C.
--
Christopher J. Fonnesbeck ( c h r i s @ f o n n e s b e c k . o r g )
Georgia Cooperative Fish & Wildlife Research Unit, University of Georgia


From pearu at scipy.org  Tue Feb 17 10:47:48 2004
From: pearu at scipy.org (Pearu Peterson)
Date: Tue, 17 Feb 2004 09:47:48 -0600 (CST)
Subject: [SciPy-user] fftpack fails on cvs build (OSX)
In-Reply-To: <0CB9A903-615F-11D8-86FB-000A956FDAC0@fonnesbeck.org>
Message-ID: <Pine.LNX.4.44.0402170946230.26316-100000@scipy.org>



On Tue, 17 Feb 2004, Christopher Fonnesbeck wrote:

> 
> On Feb 17, 2004, at 10:26 AM, Pearu Peterson wrote:
> 
> > Could you update cvs and try again? I have added Bob's suggestion 
> > there.
> > Let me know if it works or not.
> 
> You added:
> 
> opt.append('-framework','Python')

Oops. I meant what's below.

> which gives an error, since append only takes 1 argument. Substituting:
> 
> opt.append("-framework Python")
> 
> gives the following error:
> 
> g77: Python: No such file or directory

Could you try:

opt.extend(['-framework','Python'])

?

Pearu


From chris at fonnesbeck.org  Tue Feb 17 11:35:12 2004
From: chris at fonnesbeck.org (Christopher Fonnesbeck)
Date: Tue, 17 Feb 2004 11:35:12 -0500
Subject: [SciPy-user] fftpack fails on cvs build (OSX)
In-Reply-To: <Pine.LNX.4.44.0402170946230.26316-100000@scipy.org>
References: <Pine.LNX.4.44.0402170946230.26316-100000@scipy.org>
Message-ID: <41C9C8E1-6167-11D8-86FB-000A956FDAC0@fonnesbeck.org>


On Feb 17, 2004, at 10:47 AM, Pearu Peterson wrote:

>>
>> On Feb 17, 2004, at 10:26 AM, Pearu Peterson wrote:
>>
>>> Could you update cvs and try again? I have added Bob's suggestion
>>> there.
>>> Let me know if it works or not.
>>
>> You added:
>>
>> opt.append('-framework','Python')
>
> Oops. I meant what's below.
>
>> which gives an error, since append only takes 1 argument.  
>> Substituting:
>>
>> opt.append("-framework Python")
>>
>> gives the following error:
>>
>> g77: Python: No such file or directory
>
> Could you try:
>
> opt.extend(['-framework','Python'])
>

Doesnt work ... it can't seem to find python. I believe the appropriate  
way to link frameworks with g77 with "-Wl,-framework -Wl,Python". Using

opt.append("-Wl,-framework -Wl,Python -lcc_dynamic")

I am able to eliminate most of the errors, except for the following:

/usr/local/bin/g77 -Wl,-framework -Wl,Python -lcc_dynamic  
build/temp.darwin-7.2.0-Power_Macintosh-2.3/build/src/Lib/fftpack/ 
_fftpackmodule.o  
build/temp.darwin-7.2.0-Power_Macintosh-2.3/Lib/fftpack/src/zfft.o  
build/temp.darwin-7.2.0-Power_Macintosh-2.3/Lib/fftpack/src/drfft.o  
build/temp.darwin-7.2.0-Power_Macintosh-2.3/Lib/fftpack/src/zrfft.o  
build/temp.darwin-7.2.0-Power_Macintosh-2.3/Lib/fftpack/src/zfftnd.o  
build/temp.darwin-7.2.0-Power_Macintosh-2.3/build/src/fortranobject.o  
-L/usr/local/lib -L/usr/local/lib/gcc/powerpc-apple-darwin7.0.0/3.4  
-Lbuild/temp.darwin-7.2.0-Power_Macintosh-2.3 -ldfftpack -lrfftw -lfftw  
-lg2c -o  
build/lib.darwin-7.2.0-Power_Macintosh-2.3/scipy/fftpack/_fftpack.so
/usr/bin/ld: Undefined symbols:
_MAIN__
collect2: ld returned 1 exit status
error: Command "/usr/local/bin/g77 -Wl,-framework -Wl,Python  
-lcc_dynamic  
build/temp.darwin-7.2.0-Power_Macintosh-2.3/build/src/Lib/fftpack/ 
_fftpackmodule.o  
build/temp.darwin-7.2.0-Power_Macintosh-2.3/Lib/fftpack/src/zfft.o  
build/temp.darwin-7.2.0-Power_Macintosh-2.3/Lib/fftpack/src/drfft.o  
build/temp.darwin-7.2.0-Power_Macintosh-2.3/Lib/fftpack/src/zrfft.o  
build/temp.darwin-7.2.0-Power_Macintosh-2.3/Lib/fftpack/src/zfftnd.o  
build/temp.darwin-7.2.0-Power_Macintosh-2.3/build/src/fortranobject.o  
-L/usr/local/lib -L/usr/local/lib/gcc/powerpc-apple-darwin7.0.0/3.4  
-Lbuild/temp.darwin-7.2.0-Power_Macintosh-2.3 -ldfftpack -lrfftw -lfftw  
-lg2c -o  
build/lib.darwin-7.2.0-Power_Macintosh-2.3/scipy/fftpack/_fftpack.so"  
failed with exit status 1

--
Christopher J. Fonnesbeck ( c h r i s @ f o n n e s b e c k . o r g )
Georgia Cooperative Fish & Wildlife Research Unit, University of Georgia


From bob at redivi.com  Tue Feb 17 11:51:46 2004
From: bob at redivi.com (Bob Ippolito)
Date: Tue, 17 Feb 2004 11:51:46 -0500
Subject: [SciPy-user] fftpack fails on cvs build (OSX)
In-Reply-To: <41C9C8E1-6167-11D8-86FB-000A956FDAC0@fonnesbeck.org>
References: <Pine.LNX.4.44.0402170946230.26316-100000@scipy.org>
	<41C9C8E1-6167-11D8-86FB-000A956FDAC0@fonnesbeck.org>
Message-ID: <9279BE5E-6169-11D8-B0CA-000A95686CD8@redivi.com>


On Feb 17, 2004, at 11:35 AM, Christopher Fonnesbeck wrote:

>
> On Feb 17, 2004, at 10:47 AM, Pearu Peterson wrote:
>
>>>
>>> On Feb 17, 2004, at 10:26 AM, Pearu Peterson wrote:
>>>
>>>> Could you update cvs and try again? I have added Bob's suggestion
>>>> there.
>>>> Let me know if it works or not.
>>>
>>> You added:
>>>
>>> opt.append('-framework','Python')
>>
>> Oops. I meant what's below.
>>
>>> which gives an error, since append only takes 1 argument.  
>>> Substituting:
>>>
>>> opt.append("-framework Python")
>>>
>>> gives the following error:
>>>
>>> g77: Python: No such file or directory
>>
>> Could you try:
>>
>> opt.extend(['-framework','Python'])
>>
>
> Doesnt work ... it can't seem to find python. I believe the  
> appropriate way to link frameworks with g77 with "-Wl,-framework  
> -Wl,Python". Using
>
> opt.append("-Wl,-framework -Wl,Python -lcc_dynamic")
>
> I am able to eliminate most of the errors, except for the following:
>
> /usr/local/bin/g77 -Wl,-framework -Wl,Python -lcc_dynamic  
> build/temp.darwin-7.2.0-Power_Macintosh-2.3/build/src/Lib/fftpack/ 
> _fftpackmodule.o  
> build/temp.darwin-7.2.0-Power_Macintosh-2.3/Lib/fftpack/src/zfft.o  
> build/temp.darwin-7.2.0-Power_Macintosh-2.3/Lib/fftpack/src/drfft.o  
> build/temp.darwin-7.2.0-Power_Macintosh-2.3/Lib/fftpack/src/zrfft.o  
> build/temp.darwin-7.2.0-Power_Macintosh-2.3/Lib/fftpack/src/zfftnd.o  
> build/temp.darwin-7.2.0-Power_Macintosh-2.3/build/src/fortranobject.o  
> -L/usr/local/lib -L/usr/local/lib/gcc/powerpc-apple-darwin7.0.0/3.4  
> -Lbuild/temp.darwin-7.2.0-Power_Macintosh-2.3 -ldfftpack -lrfftw  
> -lfftw -lg2c -o  
> build/lib.darwin-7.2.0-Power_Macintosh-2.3/scipy/fftpack/_fftpack.so
> /usr/bin/ld: Undefined symbols:
> _MAIN__

It probably needs a -bundle, right now it's probably trying to link as  
an executable..

-bob


From chris at fonnesbeck.org  Tue Feb 17 12:17:42 2004
From: chris at fonnesbeck.org (Christopher Fonnesbeck)
Date: Tue, 17 Feb 2004 12:17:42 -0500
Subject: [SciPy-user] fftpack fails on cvs build (OSX)
In-Reply-To: <9279BE5E-6169-11D8-B0CA-000A95686CD8@redivi.com>
References: <Pine.LNX.4.44.0402170946230.26316-100000@scipy.org>
	<41C9C8E1-6167-11D8-86FB-000A956FDAC0@fonnesbeck.org>
	<9279BE5E-6169-11D8-B0CA-000A95686CD8@redivi.com>
Message-ID: <32027C9A-616D-11D8-86FB-000A956FDAC0@fonnesbeck.org>


On Feb 17, 2004, at 11:51 AM, Bob Ippolito wrote:
>>>
>>
>> Doesnt work ... it can't seem to find python. I believe the  
>> appropriate way to link frameworks with g77 with "-Wl,-framework  
>> -Wl,Python". Using
>>
>> opt.append("-Wl,-framework -Wl,Python -lcc_dynamic")
>>
>> I am able to eliminate most of the errors, except for the following:
>>
>> /usr/local/bin/g77 -Wl,-framework -Wl,Python -lcc_dynamic  
>> build/temp.darwin-7.2.0-Power_Macintosh-2.3/build/src/Lib/fftpack/ 
>> _fftpackmodule.o  
>> build/temp.darwin-7.2.0-Power_Macintosh-2.3/Lib/fftpack/src/zfft.o  
>> build/temp.darwin-7.2.0-Power_Macintosh-2.3/Lib/fftpack/src/drfft.o  
>> build/temp.darwin-7.2.0-Power_Macintosh-2.3/Lib/fftpack/src/zrfft.o  
>> build/temp.darwin-7.2.0-Power_Macintosh-2.3/Lib/fftpack/src/zfftnd.o  
>> build/temp.darwin-7.2.0-Power_Macintosh-2.3/build/src/fortranobject.o  
>> -L/usr/local/lib -L/usr/local/lib/gcc/powerpc-apple-darwin7.0.0/3.4  
>> -Lbuild/temp.darwin-7.2.0-Power_Macintosh-2.3 -ldfftpack -lrfftw  
>> -lfftw -lg2c -o  
>> build/lib.darwin-7.2.0-Power_Macintosh-2.3/scipy/fftpack/_fftpack.so
>> /usr/bin/ld: Undefined symbols:
>> _MAIN__
>
> It probably needs a -bundle, right now it's probably trying to link as  
> an executable..
>
> -bob

Right, that would do it. This appears to work:

if sys.platform=='darwin':
          
	opt.extend(["-Wl,-framework","-Wl,Python","-lcc_dynamic","-bundle"])

--
Christopher J. Fonnesbeck ( c h r i s @ f o n n e s b e c k . o r g )
Georgia Cooperative Fish & Wildlife Research Unit, University of Georgia


From chris at fonnesbeck.org  Tue Feb 17 13:22:12 2004
From: chris at fonnesbeck.org (Christopher Fonnesbeck)
Date: Tue, 17 Feb 2004 13:22:12 -0500
Subject: [SciPy-user] building fortran extensions in scipy
Message-ID: <34AD4702-6176-11D8-86FB-000A956FDAC0@fonnesbeck.org>

As I indicated in a previous post, I am trying to move some of the 
distributions pdf and likelihood methods to FORTRAN in order to speed 
them up for use in statistical simulation procedures (such as MCMC). I 
have put together one such extension, and added it to the list of 
modules in stats/setup_stats.py, as follows:

	# add futil and fdist modules
	sources = ['futil.f','fdist.f']
	sources = [os.path.join(local_path,x) for x in sources]
	for x in sources:
		ext = Extension(dot_join(parent_package,package,x),sources)
		config['ext_modules'].append(ext)

However, when I build scipy, I see that neither futil.so or fdist.so 
exist in the scipy installation. How can I persuade scipy to install 
them?

Thanks,
Chris
--
Christopher J. Fonnesbeck ( c h r i s @ f o n n e s b e c k . o r g )
Georgia Cooperative Fish & Wildlife Research Unit, University of Georgia


From chris at fonnesbeck.org  Tue Feb 17 14:11:07 2004
From: chris at fonnesbeck.org (Christopher Fonnesbeck)
Date: Tue, 17 Feb 2004 14:11:07 -0500
Subject: [SciPy-user] building fortran extensions in scipy
In-Reply-To: <34AD4702-6176-11D8-86FB-000A956FDAC0@fonnesbeck.org>
References: <34AD4702-6176-11D8-86FB-000A956FDAC0@fonnesbeck.org>
Message-ID: <0A2226E7-617D-11D8-86FB-000A956FDAC0@fonnesbeck.org>


On Feb 17, 2004, at 1:22 PM, Christopher Fonnesbeck wrote:

> As I indicated in a previous post, I am trying to move some of the 
> distributions pdf and likelihood methods to FORTRAN in order to speed 
> them up for use in statistical simulation procedures (such as MCMC). I 
> have put together one such extension, and added it to the list of 
> modules in stats/setup_stats.py, as follows:
>
> 	# add futil and fdist modules
> 	sources = ['futil.f','fdist.f']
> 	sources = [os.path.join(local_path,x) for x in sources]
> 	for x in sources:
> 		ext = Extension(dot_join(parent_package,package,x),sources)
> 		config['ext_modules'].append(ext)
>
> However, when I build scipy, I see that neither futil.so or fdist.so 
> exist in the scipy installation. How can I persuade scipy to install 
> them?
>

Ignore this ... I see my mistake. Sorry.

--
Christopher J. Fonnesbeck ( c h r i s @ f o n n e s b e c k . o r g )
Georgia Cooperative Fish & Wildlife Research Unit, University of Georgia


From pearu at scipy.org  Wed Feb 18 03:53:00 2004
From: pearu at scipy.org (Pearu Peterson)
Date: Wed, 18 Feb 2004 02:53:00 -0600 (CST)
Subject: [SciPy-user] fftpack fails on cvs build (OSX)
In-Reply-To: <32027C9A-616D-11D8-86FB-000A956FDAC0@fonnesbeck.org>
Message-ID: <Pine.LNX.4.44.0402180252001.23093-100000@scipy.org>



On Tue, 17 Feb 2004, Christopher Fonnesbeck wrote:

> Right, that would do it. This appears to work:
> 
> if sys.platform=='darwin':
>           
> 	opt.extend(["-Wl,-framework","-Wl,Python","-lcc_dynamic","-bundle"])

Thanks, it is in CVS now.

Pearu


From Jonathan.Peirce at nottingham.ac.uk  Wed Feb 11 12:55:09 2004
From: Jonathan.Peirce at nottingham.ac.uk (Jon Peirce)
Date: Wed, 11 Feb 2004 17:55:09 +0000
Subject: [SciPy-user] PIL and scipy conflicting
Message-ID: <402A6C7D.40604@psychology.nottingham.ac.uk>

Hi all

I have scipy (0.2.0_alpha_212.4128) and PIL (1.1.4) installed on a winXP 
machine and I'm getting a strange conflict. If scipy is imported second 
(using a clean namespace) then Image can no longer 'open':

#this works fine
import scipy
import Image
im = Image.open('somefile.jpg')

#this returns error
import Image
import scipy
im = Image.open('somefile.jpg')

>   File "C:\Python23\lib\site-packages\PIL\Image.py", line 1571, in open
>     raise IOError("cannot identify image file")
> IOError: cannot identify image file


Does anyone know why? I don't want just to go with the former - its 
presumably doing some other damage to scipy that i haven't come across yet!

-- 
Jon Peirce
Nottingham University
+44 (0)115 8467176 (tel)
+44 (0)115 9515324 (fax)

http://www.psychology.nottingham.ac.uk/staff/jwp/


From pearu at cens.ioc.ee  Thu Feb 19 07:01:28 2004
From: pearu at cens.ioc.ee (Pearu Peterson)
Date: Thu, 19 Feb 2004 14:01:28 +0200 (EET)
Subject: [SciPy-user] scipy OS X compiling on a G3
In-Reply-To: <31850D8C-5B31-11D8-8720-000A95843714@yale.edu>
Message-ID: <Pine.LNX.4.21.0402191348510.3832-100000@cens.kybi>


Hi all,

OSX users, could you try out the recent scipy_distutils for building
linalg module? It now uses '-framework Accelerate ..' stuff for
atlas_info but I cannot test this myself. Here are basic steps:

Install new scipy_core:

  cd scipy/scipy_core
  python setup.py install

Build linalg module:

  cd scipy/Lib/linalg
  # undefine ATLAS environment variable if any
  python setup_linalg.py build
  # pay attention to the availability of `lapack_opt_info`
  python tests/test_basic.py
  ..

Let me know if it works or not.

Note that building scipy will not work yet without atlas libraries
as scipy.integrate still uses old hooks, if someone get's scipy.linalg
buildable under darwin, I'll finish scipy support for
darwin-without-atlas asap.

Regards,
Pearu


From covino at mi.astro.it  Thu Feb 19 13:28:46 2004
From: covino at mi.astro.it (Stefano Covino)
Date: Thu, 19 Feb 2004 19:28:46 +0100
Subject: [SciPy-user] function minization with constraints
Message-ID: <1077215326.6534.42.camel@sirrah.merate.mi.astro.it>

Dear all,

is there in the SciPy package a library to minime a multi-variable
function with constraints on each parameter?

Thanks,
        Stefano



From jose.martin at wanadoo.es  Thu Feb 19 17:50:56 2004
From: jose.martin at wanadoo.es (=?iso-8859-1?Q?Jos=E9_A_Mart=EDn_H?=)
Date: Thu, 19 Feb 2004 23:50:56 +0100
Subject: [SciPy-user] PIL and scipy conflicting
References: <402A6C7D.40604@psychology.nottingham.ac.uk>
Message-ID: <007201c3f73a$d72efad0$ea93253e@CASA>

Yes thats true , this error is ...
a long time ago in a scipy faraway...
but today this error still happend...

I as you import first scipy , and then Image...



----- Original Message -----
From: "Jon Peirce" <Jonathan.Peirce at nottingham.ac.uk>
To: <scipy-user at scipy.net>
Sent: Wednesday, February 11, 2004 6:55 PM
Subject: [SciPy-user] PIL and scipy conflicting


> Hi all
>
> I have scipy (0.2.0_alpha_212.4128) and PIL (1.1.4) installed on a winXP
> machine and I'm getting a strange conflict. If scipy is imported second
> (using a clean namespace) then Image can no longer 'open':
>
> #this works fine
> import scipy
> import Image
> im = Image.open('somefile.jpg')
>
> #this returns error
> import Image
> import scipy
> im = Image.open('somefile.jpg')
>
> >   File "C:\Python23\lib\site-packages\PIL\Image.py", line 1571, in open
> >     raise IOError("cannot identify image file")
> > IOError: cannot identify image file
>
>
> Does anyone know why? I don't want just to go with the former - its
> presumably doing some other damage to scipy that i haven't come across
yet!
>
> --
> Jon Peirce
> Nottingham University
> +44 (0)115 8467176 (tel)
> +44 (0)115 9515324 (fax)
>
> http://www.psychology.nottingham.ac.uk/staff/jwp/
>
> _______________________________________________
> SciPy-user mailing list
> SciPy-user at scipy.net
> http://www.scipy.net/mailman/listinfo/scipy-user


From nwagner at mecha.uni-stuttgart.de  Fri Feb 20 08:13:25 2004
From: nwagner at mecha.uni-stuttgart.de (Nils Wagner)
Date: Fri, 20 Feb 2004 14:13:25 +0100
Subject: [SciPy-user] Urgent: Bug in linalg.lu_factor linalg.lu_solve 
Message-ID: <403607F5.C043379@mecha.uni-stuttgart.de>

Hi all,

I have observed a very strange behaviour within linalg.lu_factor /
linalg.lu_solve.
Any idea or suggestion ?

Nils

from scipy import *

def F(A,l):

   tmp=zeros((n,n),Complex)
   tmp[:n,:n] = A+l*identity(n)+exp(-l)*identity(n)
   return tmp

def Fs(l):

   tmp=zeros((n,n),Complex)
   tmp[:n,:n] = identity(n)-l*exp(-l)*identity(n)
   return tmp

def T(n):

   tmp=zeros((n,n),Complex)
   for i in arange(0,n):
     tmp[i,i] = -2.0
     if i < n-1:
       tmp[i+1,i] = 1.0
       tmp[i,i+1] = 1.0
   return tmp

n = 4
A = T(n)

gamma = 2.0+0j
rho = 2.0

ns = 180
#for i in arange(0,ns):
for i in arange(0,2):
   l  = gamma+rho*exp(2.*pi*i*1j/ns)
   D1 = F(A,l)
   Ds = Fs(l)

   Dinv = linalg.inv(D1)
   lu, piv = linalg.lu_factor(D1,overwrite_a=0)

   tr = 0.0
   for k in arange(0,n):
      rhs = Ds[:,k]
      g1  = linalg.lu_solve((lu,piv),rhs,trans=0,overwrite_b=0)
      tr  = tr + g1[k]

#
#  tr and trace(dot(Dinv,Ds)) should be the same !!!!!
#
   print tr, trace(dot(Dinv,Ds))


From nwagner at mecha.uni-stuttgart.de  Fri Feb 20 08:19:01 2004
From: nwagner at mecha.uni-stuttgart.de (Nils Wagner)
Date: Fri, 20 Feb 2004 14:19:01 +0100
Subject: [SciPy-user] Bug in linalg.lu_factor linalg.lu_solve 
Message-ID: <40360945.8C842433@mecha.uni-stuttgart.de>

Hi all,

the problem with linalg.lu_factor and linalg.lu_solve
seems to be restricted to complex matrices...

Nils


From nwagner at mecha.uni-stuttgart.de  Fri Feb 20 08:56:30 2004
From: nwagner at mecha.uni-stuttgart.de (Nils Wagner)
Date: Fri, 20 Feb 2004 14:56:30 +0100
Subject: [SciPy-user] linalg.lu is 
Message-ID: <4036120E.5B9DE9B8@mecha.uni-stuttgart.de>

Hi all,

linalg.lu cannot handle complex equations, where the rhs is complex.

Nils

from scipy import *
from RandomArray import *
n = 2
A = rand(n,n)+rand(n,n)*1j
X = zeros((n,n),Complex)
Ainv = linalg.inv(A)
print linalg.norm(dot(Ainv,A))
I = identity(n)+0j*identity(n)
#
# linalg.solve cannot handle complex rhs !!
#
#I = I.real
#
for i in arange(0,n):
   X[:,i] = linalg.solve(A,I[:,i])
#
# Should be the inverse of A as well !!!!
#
print X
#
# Inverse of A
#
print Ainv


From elcortostrash at gmx.net  Fri Feb 20 11:01:04 2004
From: elcortostrash at gmx.net (el corto)
Date: Fri, 20 Feb 2004 17:01:04 +0100 (MET)
Subject: [SciPy-user] install SciPy with cygwin and Python 2.3x
Message-ID: <23254.1077292864@www45.gmx.net>

I've downloaded and installed the latest version of cygwin which comes with
Python 2.3.3. The install instructions at 

   www.scipy.org/site_content/tutorials/install_binaries 

say to unpack the SciPy-binary for cygwin in 

   /usr/lib/python2.3/site-packages/

which I did. The binary includes the folders

scipy
gui_thread
scipy_base
scipy_distutils

Since I'm new to the cygwin thing I really don't know if that was the whole
installation (sry for this obviously stupid question).

When I try to

   >>> import scipy

I get error-messages like

    File "/usr/lib/python2.3/site-packages/scipy/__init__.py", line 36, in ?
          from scipy_base import *
    File "/usr/lib/python2.3/site-packages/scipy_base/__init__.py", line
110, in ?
          raise ImportError, mess
    ImportError: dlopen: Win32 error 126

OK, now what does that mean ?? I really hope that someone can give me a
little hint :-) 

thx, bye.
 

-- 
GMX ProMail (250 MB Mailbox, 50 FreeSMS, Virenschutz, 2,99 EUR/Monat...)
jetzt 3 Monate GRATIS + 3x DER SPIEGEL +++ http://www.gmx.net/derspiegel +++


From pearu at scipy.org  Fri Feb 20 11:40:49 2004
From: pearu at scipy.org (Pearu Peterson)
Date: Fri, 20 Feb 2004 10:40:49 -0600 (CST)
Subject: [SciPy-user] install SciPy with cygwin and Python 2.3x
In-Reply-To: <23254.1077292864@www45.gmx.net>
Message-ID: <Pine.LNX.4.44.0402201036350.25361-100000@scipy.org>



On Fri, 20 Feb 2004, el corto wrote:

> I've downloaded and installed the latest version of cygwin which comes with
> Python 2.3.3. The install instructions at 
> 
>    www.scipy.org/site_content/tutorials/install_binaries 
> 
<snip>

> I get error-messages like
> 
>     File "/usr/lib/python2.3/site-packages/scipy/__init__.py", line 36, in ?
>           from scipy_base import *
>     File "/usr/lib/python2.3/site-packages/scipy_base/__init__.py", line
> 110, in ?
>           raise ImportError, mess
>     ImportError: dlopen: Win32 error 126
> 
> OK, now what does that mean ?? I really hope that someone can give me a
> little hint :-) 

I cannot say what causes the above error but I suggest building scipy
under cygwin yourself, it is not so hard. See

  http://cens.ioc.ee/~pearu/scipy/BUILD_WIN32.html

HTH,
Pearu


From oliphant at ee.byu.edu  Fri Feb 20 11:37:28 2004
From: oliphant at ee.byu.edu (Travis E. Oliphant)
Date: Fri, 20 Feb 2004 09:37:28 -0700
Subject: [SciPy-user] PIL and scipy conflicting
In-Reply-To: <402A6C7D.40604@psychology.nottingham.ac.uk>
References: <402A6C7D.40604@psychology.nottingham.ac.uk>
Message-ID: <403637C8.7030409@ee.byu.edu>

Jon Peirce wrote:
> Hi all
> 
> I have scipy (0.2.0_alpha_212.4128) and PIL (1.1.4) installed on a winXP 
> machine and I'm getting a strange conflict. If scipy is imported second 
> (using a clean namespace) then Image can no longer 'open':
> 

You may be interested to know that if you have PIL installed then scipy 
has some commands that make opening and saving images to and from 
Numeric arrays even easier.  I have used these commands without incident .

arr = imread('somefile.jpg')
imsave('newfile.jpg',arr)

One big difference is that the image is read in to a Numeric array 
(which of course an image should be thought of as anyway :-) )

Best,

Travis O.



From pearu at scipy.org  Fri Feb 20 12:12:33 2004
From: pearu at scipy.org (Pearu Peterson)
Date: Fri, 20 Feb 2004 11:12:33 -0600 (CST)
Subject: [SciPy-user] linalg.lu is 
In-Reply-To: <4036120E.5B9DE9B8@mecha.uni-stuttgart.de>
Message-ID: <Pine.LNX.4.44.0402201107590.25361-100000@scipy.org>



On Fri, 20 Feb 2004, Nils Wagner wrote:

> linalg.lu cannot handle complex equations, where the rhs is complex.

This is not generally true but you are right, something goes wrong.
For example,

In [4]: from scipy.linalg import *

In [5]: a = [[1,1],[1,0]] # ok

In [6]: x0 = solve(a,[1,0j])

In [7]: dot(a,x0)
Out[7]: array([ 1.+0.j,  0.+0.j])

In [8]: a = [[1,1],[1.2,0]] # not ok

In [9]: x0 = solve(a,[1,0j])

In [10]: dot(a,x0)
Out[10]: array([ 0.83333333+0.j,  0.        +0.j])

When I disable ATLAS, then the second case gives correct answer.
So, I suspect it is an ATLAS bug. I am using 3.6.0. 
What version of ATLAS are you using?

Pearu


From dr.nwagner at web.de  Fri Feb 20 13:28:08 2004
From: dr.nwagner at web.de (Nils Wagner)
Date: Fri, 20 Feb 2004 19:28:08 +0100
Subject: [SciPy-user] linalg.lu is 
Message-ID: <200402201828.i1KIS8Q32517@mailgate5.cinetic.de>

SciPy Users List <scipy-user at scipy.net> schrieb am 20.02.04 18:06:43: 
 
 
 
On Fri, 20 Feb 2004, Nils Wagner wrote: 
 
> linalg.lu cannot handle complex equations, where the rhs is complex. 
 
This is not generally true but you are right, something goes wrong. 
For example, 
 
In [4]: from scipy.linalg import * 
 
In [5]: a = [[1,1],[1,0]] # ok 
 
In [6]: x0 = solve(a,[1,0j]) 
 
In [7]: dot(a,x0) 
Out[7]: array([ 1.+0.j,  0.+0.j]) 
 
In [8]: a = [[1,1],[1.2,0]] # not ok 
 
In [9]: x0 = solve(a,[1,0j]) 
 
In [10]: dot(a,x0) 
Out[10]: array([ 0.83333333+0.j,  0.        +0.j]) 
 
When I disable ATLAS, then the second case gives correct answer. 
So, I suspect it is an ATLAS bug. I am using 3.6.0.  
What version of ATLAS are you using? 
 
I am using 3.6 as well. How do I disable the usage of ATLAS ? 
Did you sent an email to the maintainer of ATLAS ? 
 
Nils 
 
Pearu 
 
_______________________________________________ 
SciPy-user mailing list 
SciPy-user at scipy.net 
http://www.scipy.net/mailman/listinfo/scipy-user 
 
______________________________________________________________________________
Ein Grund zum Feiern: Die PC Praxis ermittelt zwischen 10 grossen
Mailprovidern WEB.DE FreeMail als Testsieger http://f.web.de/?mc=021190


From dr.nwagner at web.de  Sat Feb 21 01:37:18 2004
From: dr.nwagner at web.de (Nils Wagner)
Date: Sat, 21 Feb 2004 07:37:18 +0100
Subject: [SciPy-user] Disable ATLAS
Message-ID: <200402210637.i1L6bIQ32423@mailgate5.cinetic.de>

Hi all, 
 
How can I disable the usage of ATLAS in scipy ? 
 
Nils 
 
______________________________________________________________________________
Nachrichten, Musik und Spiele schnell und einfach per Quickstart im 
WEB.DE Screensaver - Gratis downloaden: http://screensaver.web.de/?mc=021110


From pearu at scipy.org  Sat Feb 21 01:52:46 2004
From: pearu at scipy.org (Pearu Peterson)
Date: Sat, 21 Feb 2004 00:52:46 -0600 (CST)
Subject: [SciPy-user] Disable ATLAS
In-Reply-To: <200402210637.i1L6bIQ32423@mailgate5.cinetic.de>
Message-ID: <Pine.LNX.4.44.0402210047400.25361-100000@scipy.org>



On Sat, 21 Feb 2004, Nils Wagner wrote:

> Hi all, 
>  
> How can I disable the usage of ATLAS in scipy ? 

There are several ways to do it, all are hackish currently.
E.g. set

  atlas_info={}

in lapack_opt_info.calc_info method in system_info.py file
or hide the location of atlas libraries from system_info.py.
Then you'll need Fortran blas and lapack installed.

But considering the current issue, you can either try out
different ALTAS versions or set
  force_clapack = 0
in the get_lapack_funcs() argument list. See linalg/lapack.py file.

HTH,
Pearu


From dr.nwagner at web.de  Sat Feb 21 08:27:17 2004
From: dr.nwagner at web.de (Nils Wagner)
Date: Sat, 21 Feb 2004 14:27:17 +0100
Subject: [SciPy-user] Disable ATLAS
Message-ID: <200402211327.i1LDRGQ23314@mailgate5.cinetic.de>

SciPy Users List <scipy-user at scipy.net> schrieb am 21.02.04 07:46:32: 
 
 
 
On Sat, 21 Feb 2004, Nils Wagner wrote: 
 
> Hi all,  
>   
> How can I disable the usage of ATLAS in scipy ?  
 
There are several ways to do it, all are hackish currently. 
E.g. set 
 
  atlas_info={} 
 
in lapack_opt_info.calc_info method in system_info.py file 
or hide the location of atlas libraries from system_info.py. 
Then you'll need Fortran blas and lapack installed. 
 
But considering the current issue, you can either try out 
different ALTAS versions or set 
  force_clapack = 0 
in the get_lapack_funcs() argument list. See linalg/lapack.py file. 
 
I have disabled ATLAS. The problem with linalg.lu didn't vanish. 
What can we do ? 
 
 
from scipy import * 
from RandomArray import * 
n = 2 
A = rand(n,n)+rand(n,n)*1j 
X = zeros((n,n),Complex) 
Ainv = linalg.inv(A) 
print linalg.norm(dot(Ainv,A)) 
I = identity(n)+identity(n)*0j 
# 
# linalg.solve cannot handle complex rhs !! 
# 
#I = I.real 
# 
for i in arange(0,n): 
   r = I[:,i] 
   X[:,i] = linalg.solve(A,r) 
# 
# Should be the inverse of A as well !!!! 
# 
print 
print 'X' 
print 
print X 
# 
# Inverse of A 
# 
print 
print 'inv(A)' 
print 
print Ainv 
 
print dot(A,Ainv) 
print dot(A,X) 
 
 
HTH, 
Pearu 
 
_______________________________________________ 
SciPy-user mailing list 
SciPy-user at scipy.net 
http://www.scipy.net/mailman/listinfo/scipy-user 
 
______________________________________________________________________________
... and the winner is... WEB.DE FreeMail! - Deutschlands beste E-Mail
ist zum 39. Mal Testsieger (PC Praxis 03/04) http://f.web.de/?mc=021191


From pearu at scipy.org  Sat Feb 21 08:40:56 2004
From: pearu at scipy.org (Pearu Peterson)
Date: Sat, 21 Feb 2004 07:40:56 -0600 (CST)
Subject: [SciPy-user] Disable ATLAS
In-Reply-To: <200402211327.i1LDRGQ23314@mailgate5.cinetic.de>
Message-ID: <Pine.LNX.4.44.0402210737230.25361-100000@scipy.org>



On Sat, 21 Feb 2004, Nils Wagner wrote:

> I have disabled ATLAS. The problem with linalg.lu didn't vanish. 
> What can we do ? 

Are you absolutely sure that ATLAS is not being used (rebuilded linalg 
from scratch, removed earlier installations, eg `ldd blas.so` does not 
contain references to ATLAS symbols, etc etc)?
Have you tried the two examples of mine from the earlier thread?

Pearu


From dr.nwagner at web.de  Sat Feb 21 12:32:15 2004
From: dr.nwagner at web.de (Nils Wagner)
Date: Sat, 21 Feb 2004 18:32:15 +0100
Subject: [SciPy-user] Disable ATLAS
Message-ID: <200402211732.i1LHWFQ13708@mailgate5.cinetic.de>

SciPy Users List <scipy-user at scipy.net> schrieb am 21.02.04 14:35:12: 
 
 
 
On Sat, 21 Feb 2004, Nils Wagner wrote: 
 
> I have disabled ATLAS. The problem with linalg.lu didn't vanish.  
> What can we do ?  
 
Are you absolutely sure that ATLAS is not being used (rebuilded linalg  
from scratch, removed earlier installations, eg `ldd blas.so` does not  
contain references to ATLAS symbols, etc etc)? 
 
Yes.  
 
nwagner at linux:/usr/local/lib/python2.3/site-packages/scipy/linalg> ldd cblas.so              
libg2c.so.0 => /usr/lib/libg2c.so.0 (0x40024000) 
        libm.so.6 => /lib/i686/libm.so.6 (0x40042000) 
        libgcc_s.so.1 => /lib/libgcc_s.so.1 (0x40065000) 
        libc.so.6 => /lib/i686/libc.so.6 (0x4006d000) 
        /lib/ld-linux.so.2 => /lib/ld-linux.so.2 (0x80000000) 
 
Have you tried the two examples of mine from the earlier thread? 
Yes. they work fine for me. but lu.py failed again 
Please can you run lu.py  
 
from scipy import * 
from RandomArray import * 
n = 2 
A = rand(n,n)+rand(n,n)*1j 
X = zeros((n,n),Complex) 
Ainv = linalg.inv(A) 
R = identity(n)+identity(n)*0j 
# 
# linalg.solve cannot handle complex rhs !! 
# 
#R = R.real 
# 
for i in arange(0,n): 
   r = R[:,i] 
   X[:,i] = linalg.solve(A,r) 
# 
# Should be the inverse of A as well !!!! 
# 
print 
print 'X' 
print 
print X 
# 
# Inverse of A 
# 
print 
print 'inv(A)' 
print 
print Ainv 
print 
print 'Identity ???' 
print 
print dot(A,Ainv) 
print 
print 'Identity ???' 
print 
print dot(A,X) 
 
Thanks in advance 
 
 Nils 
 
Pearu 
 
_______________________________________________ 
SciPy-user mailing list 
SciPy-user at scipy.net 
http://www.scipy.net/mailman/listinfo/scipy-user 
 
______________________________________________________________________________
... and the winner is... WEB.DE FreeMail! - Deutschlands beste E-Mail
ist zum 39. Mal Testsieger (PC Praxis 03/04) http://f.web.de/?mc=021191


From pearu at scipy.org  Sat Feb 21 15:09:09 2004
From: pearu at scipy.org (Pearu Peterson)
Date: Sat, 21 Feb 2004 14:09:09 -0600 (CST)
Subject: [SciPy-user] Disable ATLAS
In-Reply-To: <200402211732.i1LHWFQ13708@mailgate5.cinetic.de>
Message-ID: <Pine.LNX.4.44.0402211400070.25361-100000@scipy.org>



On Sat, 21 Feb 2004, Nils Wagner wrote:

> Please can you run lu.py  

When I set

  force_clapack=0

then the output of lu.py is

---------------
X

[[ 1.28518872-0.09428482j -0.46708202-0.37104511j]
 [-0.31203795+0.19175932j  0.73496379-0.19732829j]]

inv(A)

[[ 0.91414361+0.3727972j  -0.46708202-0.37104511j]
 [-0.50936624-0.54320447j  0.73496379-0.19732829j]]

Identity ???

[[  1.00000000e+00 +2.90295132e-17j  -1.99086624e-17 -3.05202912e-17j]
 [  7.09610322e-17 +1.95427442e-16j   1.00000000e-00 +2.33103467e-18j]]

Identity ???

[[  1.00000000e+00 +1.72794721e-17j  -1.99086624e-17 -3.05202912e-17j]
 [ -2.36085023e-17 +1.00000000e+00j   1.00000000e-00 +2.33103467e-18j]]
--------------

Everything is correct up to round-off errors. So, check again that ATLAS 
is not being used. For the record, with force_lapack=1 the results are 
incorrect.

As for a possible Atlas bug, I haven't created a simple C program that
would demonstrate the bug and is necessary to file a bug report to Atlas
people. Any takers?

Pearu


From dr.nwagner at web.de  Sat Feb 21 16:57:20 2004
From: dr.nwagner at web.de (Nils Wagner)
Date: Sat, 21 Feb 2004 22:57:20 +0100
Subject: [SciPy-user] Disable ATLAS
Message-ID: <200402212157.i1LLvKQ24535@mailgate5.cinetic.de>

SciPy Users List <scipy-user at scipy.net> schrieb am 21.02.04 21:03:36: 
 
 
 
On Sat, 21 Feb 2004, Nils Wagner wrote: 
 
> Please can you run lu.py   
 
When I set 
 
  force_clapack=0 
 
then the output of lu.py is 
 
--------------- 
X 
 
[[ 1.28518872-0.09428482j -0.46708202-0.37104511j] 
 [-0.31203795+0.19175932j  0.73496379-0.19732829j]] 
 
inv(A) 
 
[[ 0.91414361+0.3727972j  -0.46708202-0.37104511j] 
 [-0.50936624-0.54320447j  0.73496379-0.19732829j]] 
 
Identity ??? 
 
[[  1.00000000e+00 +2.90295132e-17j  -1.99086624e-17 -3.05202912e-17j] 
 [  7.09610322e-17 +1.95427442e-16j   1.00000000e-00 +2.33103467e-18j]] 
 
Identity ??? 
 
[[  1.00000000e+00 +1.72794721e-17j  -1.99086624e-17 -3.05202912e-17j] 
 [ -2.36085023e-17 +1.00000000e+00j   1.00000000e-00 +2.33103467e-18j]] 
----^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^---------- 
 
How about -2.36085023e-17 +1.00000000e+00j ? 
The imaginary part should be zero ! 
 
Nils 
 
Everything is correct up to round-off errors. So, check again that ATLAS  
is not being used. For the record, with force_lapack=1 the results are  
incorrect. 
 
As for a possible Atlas bug, I haven't created a simple C program that 
would demonstrate the bug and is necessary to file a bug report to Atlas 
people. Any takers? 
 
Pearu 
 
_______________________________________________ 
SciPy-user mailing list 
SciPy-user at scipy.net 
http://www.scipy.net/mailman/listinfo/scipy-user 
 
______________________________________________________________________________
... and the winner is... WEB.DE FreeMail! - Deutschlands beste E-Mail
ist zum 39. Mal Testsieger (PC Praxis 03/04) http://f.web.de/?mc=021191


From pearu at scipy.org  Sat Feb 21 18:12:52 2004
From: pearu at scipy.org (Pearu Peterson)
Date: Sat, 21 Feb 2004 17:12:52 -0600 (CST)
Subject: [SciPy-user] Disable ATLAS
In-Reply-To: <200402212157.i1LLvKQ24535@mailgate5.cinetic.de>
Message-ID: <Pine.LNX.4.44.0402211703300.25361-100000@scipy.org>



On Sat, 21 Feb 2004, Nils Wagner wrote:

> Identity ??? 
>  
> [[  1.00000000e+00 +1.72794721e-17j  -1.99086624e-17 -3.05202912e-17j] 
>  [ -2.36085023e-17 +1.00000000e+00j   1.00000000e-00 +2.33103467e-18j]] 
> ----^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^---------- 
>  
> How about -2.36085023e-17 +1.00000000e+00j ? 
> The imaginary part should be zero ! 

Funny that I missed that, sorry. I think I found the cause of these 
incorrect results. It seems to be due to solve() ignoring piv output from 
gesv function.

Pearu


From pearu at scipy.org  Sun Feb 22 04:53:18 2004
From: pearu at scipy.org (Pearu Peterson)
Date: Sun, 22 Feb 2004 03:53:18 -0600 (CST)
Subject: [SciPy-user] Incorrect results in linalg.solve and friends
In-Reply-To: <200402211732.i1LHWFQ13708@mailgate5.cinetic.de>
Message-ID: <Pine.LNX.4.44.0402220337110.25361-100000@scipy.org>


Hi,

This is heads-up for linalg crunchers. Incorrect results appear when
the following condition holds:

  an input array is a *slice* of a complex array.

> for i in arange(0,n):

For example, changing
 
>    r = R[:,i] 

to `r = R[:,i].copy()` will give correct results.

>    X[:,i] = linalg.solve(A,r) 

This gives a workaround to avoid this problem until it gets resolved.
Also, this is not due to some bug in Atlas but rather of copy_ND_array in
fortranobject.c or of PyArray_CopyFromObject, but most probably of
copy_ND_array.

Pearu


From chris at fonnesbeck.org  Mon Feb 23 12:37:12 2004
From: chris at fonnesbeck.org (Christopher Fonnesbeck)
Date: Mon, 23 Feb 2004 12:37:12 -0500
Subject: [SciPy-user] distutils problem on cvs install under python 2.3
In-Reply-To: <Pine.LNX.4.44.0402041332090.9269-100000@scipy.org>
References: <Pine.LNX.4.44.0402041332090.9269-100000@scipy.org>
Message-ID: <E9D836D1-6626-11D8-AF75-000A956FDAC0@fonnesbeck.org>


On Feb 4, 2004, at 2:39 PM, Pearu Peterson wrote:

>
>
> On Wed, 4 Feb 2004, Christopher Fonnesbeck wrote:
>
>> When I attempt a build of the latest cvs on a clean Debian Linux
>> install of python2.3, I get this old error message:
>>
>> chris at redhead:/usr/src/scipy$ python setup.py build
>> Traceback (most recent call last):
>>    File "setup.py", line 25, in ?
>>      from scipy_distutils.core import setup
>>    File "scipy_core/scipy_distutils/core.py", line 1, in ?
>>      from distutils.core import *
>> ImportError: No module named distutils.core
>
> Could you verify that adding `import distutils` before the line #1
> fixes this import error?
>
>> I thought this bug in distutils was squashed long ago. Am I missing
>> something here?
>
> Strange, I am on debian too with Python 2.3.3 and I don't get this 
> error.
> Btw, do you have write permissions to
> /usr/src/scipy/scipy_core/scipy_distutils directory? I noticed that
> the traceback comes from .py file and not from .pyc file.
>

I have tried reinstalling python2.3 on Debian, but when I try and 
import distutils, it tells me that there is no model named distutils! 
It seems that the current testing/stable distributions of python2.3 are 
broken.

C.

--
Christopher J. Fonnesbeck ( c h r i s @ f o n n e s b e c k . o r g )
Georgia Cooperative Fish & Wildlife Research Unit, University of Georgia


From chris at fonnesbeck.org  Mon Feb 23 16:21:22 2004
From: chris at fonnesbeck.org (Christopher Fonnesbeck)
Date: Mon, 23 Feb 2004 16:21:22 -0500
Subject: IGNORE [SciPy-user] distutils problem on cvs install under python 2.3
In-Reply-To: <E9D836D1-6626-11D8-AF75-000A956FDAC0@fonnesbeck.org>
References: <Pine.LNX.4.44.0402041332090.9269-100000@scipy.org>
	<E9D836D1-6626-11D8-AF75-000A956FDAC0@fonnesbeck.org>
Message-ID: <3AE1A59C-6646-11D8-B3C2-000A956FDAC0@fonnesbeck.org>


On Feb 23, 2004, at 12:37 PM, Christopher Fonnesbeck wrote:

>
> On Feb 4, 2004, at 2:39 PM, Pearu Peterson wrote:
>
>>
>>
>> On Wed, 4 Feb 2004, Christopher Fonnesbeck wrote:
>>
>>> When I attempt a build of the latest cvs on a clean Debian Linux
>>> install of python2.3, I get this old error message:
>>>
>>> chris at redhead:/usr/src/scipy$ python setup.py build
>>> Traceback (most recent call last):
>>>    File "setup.py", line 25, in ?
>>>      from scipy_distutils.core import setup
>>>    File "scipy_core/scipy_distutils/core.py", line 1, in ?
>>>      from distutils.core import *
>>> ImportError: No module named distutils.core
>>
>> Could you verify that adding `import distutils` before the line #1
>> fixes this import error?
>>
>>> I thought this bug in distutils was squashed long ago. Am I missing
>>> something here?
>>
>> Strange, I am on debian too with Python 2.3.3 and I don't get this 
>> error.
>> Btw, do you have write permissions to
>> /usr/src/scipy/scipy_core/scipy_distutils directory? I noticed that
>> the traceback comes from .py file and not from .pyc file.
>>
>
> I have tried reinstalling python2.3 on Debian, but when I try and 
> import distutils, it tells me that there is no model named distutils! 
> It seems that the current testing/stable distributions of python2.3 
> are broken.
>

Ignore this; I didnt have the proper python-dev installed.

C.
--
Christopher J. Fonnesbeck ( c h r i s @ f o n n e s b e c k . o r g )
Georgia Cooperative Fish & Wildlife Research Unit, University of Georgia


From nwagner at mecha.uni-stuttgart.de  Mon Feb 23 02:54:01 2004
From: nwagner at mecha.uni-stuttgart.de (Nils Wagner)
Date: Mon, 23 Feb 2004 08:54:01 +0100
Subject: [SciPy-user] Incorrect results in linalg.solve and friends
References: <Pine.LNX.4.44.0402220337110.25361-100000@scipy.org>
Message-ID: <4039B199.EA7FCDFD@mecha.uni-stuttgart.de>

Pearu Peterson schrieb:
> 
> Hi,
> 
> This is heads-up for linalg crunchers. Incorrect results appear when
> the following condition holds:
> 
>   an input array is a *slice* of a complex array.
> 
> > for i in arange(0,n):
> 
> For example, changing
> 
> >    r = R[:,i]
> 
> to `r = R[:,i].copy()` will give correct results.
> 
> >    X[:,i] = linalg.solve(A,r)
> 
> This gives a workaround to avoid this problem until it gets resolved.
> Also, this is not due to some bug in Atlas but rather of copy_ND_array in
> fortranobject.c or of PyArray_CopyFromObject, but most probably of
> copy_ND_array.
> 
> Pearu

Pearu,

Afaik fortranobject.c belongs to your f2py project. Therefore, it would
be very kind
of you, if you put a new F2PY release on your website.

Thanks in advance

                     Nils
 
BTW, the cvs repository of scipy is currently not available.
> 
> _______________________________________________
> SciPy-user mailing list
> SciPy-user at scipy.net
> http://www.scipy.net/mailman/listinfo/scipy-user


From nwagner at mecha.uni-stuttgart.de  Mon Feb 23 02:41:58 2004
From: nwagner at mecha.uni-stuttgart.de (Nils Wagner)
Date: Mon, 23 Feb 2004 08:41:58 +0100
Subject: [SciPy-user] CVS repository is not availabe
Message-ID: <4039AEC6.490930D@mecha.uni-stuttgart.de>

 


From pearu at scipy.org  Mon Feb 23 18:04:36 2004
From: pearu at scipy.org (Pearu Peterson)
Date: Mon, 23 Feb 2004 17:04:36 -0600 (CST)
Subject: [SciPy-user] Incorrect results in linalg.solve and friends
In-Reply-To: <Pine.LNX.4.44.0402220337110.25361-100000@scipy.org>
Message-ID: <Pine.LNX.4.44.0402231658210.1416-100000@scipy.org>


Hi,

On Sun, 22 Feb 2004, Pearu Peterson wrote:

> This is heads-up for linalg crunchers. Incorrect results appear when
> the following condition holds:
> 
>   an input array is a *slice* of a complex array.

More accurate condition for the bug to show up is:

   an input array is an one-dimensional slice of a complex array

> This gives a workaround to avoid this problem until it gets resolved.
> Also, this is not due to some bug in Atlas but rather of copy_ND_array in
> fortranobject.c or of PyArray_CopyFromObject, but most probably of
> copy_ND_array.

This issue is now resolved. Upgrading f2py to the latest is highly
recommended, especially when computations use complex inputs to f2py
generated wrapper functions.

Regards,
Pearu


From nwagner at mecha.uni-stuttgart.de  Tue Feb 24 03:02:11 2004
From: nwagner at mecha.uni-stuttgart.de (Nils Wagner)
Date: Tue, 24 Feb 2004 09:02:11 +0100
Subject: [SciPy-user] Binomial coefficients
Message-ID: <403B0503.AE69099@mecha.uni-stuttgart.de>

Hi all,

I am wondering if binomial coefficients are currently available in
scipy.
If that is not the case, I would like to add this feature to scipy
Any comment ?

Nils


from scipy import *

def binomial(a,b):
  return
int(special.gamma(a+1)/(special.gamma(b+1)*special.gamma(a-b+1)))

#
# The following table is given in Bronstein Taschenbuch der Mathematik
#
for n in arange(0,21):
  for k in arange(0,11):
    print binomial(n,k)
  print


From rkern at ucsd.edu  Tue Feb 24 03:53:09 2004
From: rkern at ucsd.edu (Robert Kern)
Date: Tue, 24 Feb 2004 00:53:09 -0800
Subject: [SciPy-user] Binomial coefficients
In-Reply-To: <403B0503.AE69099@mecha.uni-stuttgart.de>
References: <403B0503.AE69099@mecha.uni-stuttgart.de>
Message-ID: <DEF952D8-66A6-11D8-B725-000A95D9DAF6@ucsd.edu>

On Feb 24, 2004, at 12:02 AM, Nils Wagner wrote:

> Hi all,
>
> I am wondering if binomial coefficients are currently available in
> scipy.

It's named scipy.comb()

-- 
Robert Kern
rkern at ucsd.edu
"In the fields of hell where the grass grows high
  Are the graves of dreams allowed to die."
    -- Richard Harter


From Bob.Cowdery at CGI-Europe.com  Tue Feb 24 04:41:32 2004
From: Bob.Cowdery at CGI-Europe.com (Bob.Cowdery at CGI-Europe.com)
Date: Tue, 24 Feb 2004 09:41:32 -0000
Subject: [SciPy-user] Help with filter design
Message-ID: <9B40DB5FBF7D2048AD79C2A91F862A5106DCFF@STEVAPPS05.Stevenage.CGI-Europe.com>

Forgive these questions if they are rather basic. I am trying to get up to
speed with python and the scientific packages. I have an implementation of a
dsp process for a software radio implemented using Intel IPP. I am
converting this to python. One of the things I need to do is fast
convolution filtering. To create the filter in IPP I do something like:
ippsFIRGenLowpass_64f( ( high_cutoff - low_cutoff )/2, taps, no_of_taps,
window_type ) ) where taps is the output array.

Is there something similar in SciPy or other packages. I found the tutorial
in the signal processing area too difficult to follow and I can't seem to
locate any other documentation for SciPy except the help where you really
need to know what you are looking for.

Any assistance appreciated
Bob
 
 
Bob Cowdery
CGI Senior Technical Architect
+44(0)1438 791517
Mobile: +44(0)7771 532138
bob.cowdery at cgi-europe.com <mailto:bob.cowdery at cgi-europe.com> 
 
 
 

*** Confidentiality Notice *** Proprietary/Confidential
Information belonging to CGI Group Inc. and its affiliates
may be contained in this message. If you are not a recipient
indicated or intended in this message (or responsible for
delivery of this message to such person), or you think for
any reason that this message may have been addressed to you
in error, you may not use or copy or deliver this message
to anyone else.  In such case, you should destroy this
message and are asked to notify the sender by reply email.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.scipy.org/pipermail/scipy-user/attachments/20040224/ee851a93/attachment.html>

From pearu at cens.ioc.ee  Tue Feb 24 05:02:08 2004
From: pearu at cens.ioc.ee (Pearu Peterson)
Date: Tue, 24 Feb 2004 12:02:08 +0200 (EET)
Subject: [SciPy-user] ANN: F2PY - Fortran to Python Interface Generator
Message-ID: <Pine.LNX.4.21.0402241109420.29453-100000@cens.kybi>


F2PY - Fortran to Python Interface Generator
--------------------------------------------

I am pleased to announce the seventh public release of F2PY, version
2.39.235_1642. The purpose of the F2PY project is to provide the 
connection between Python and Fortran programming languages.

For more information, see

  http://cens.ioc.ee/projects/f2py2e/

Download source:

  http://cens.ioc.ee/projects/f2py2e/2.x/F2PY-2-latest.tar.gz

What's new?
------------

Since the last public release, that was more than two years ago, F2PY
project has been actively maintained and many useful features have been
implemented. Here follows some highlights:

* New statement ``usercode`` allows inserting user defined C code to F2PY
  generated extension module sources at various relevant places: before
  wrapper functions, after variable declarations, and to the end of
  initialization function of the module. This makes using F2PY very
  flexible.

* New statement ``pymethoddef`` allows adding items to PyMethodDef-array.

* Full support for character arrays and arrays of strings is finally
  implemented.

* Number of feature requests from users are implemented. For example:
  - the auxiliary ``as_column_major_storage()`` that efficiently converts
    array to column storage order
  - the F2PY_REPORT_ON_ARRAY_COPY macro that when defined sends a message
    to stderr when copy of an array with the size larger than specified
    threshold is made.
  - Numarray support, thanks to Todd Miller.

* Support for Win32 and Mac OSX platforms is considerably improved. The
  list of compilers supported by the scipy_distutils package is longer
  than ever:
    GNU Fortran Compiler
    Portland Group Fortran Compiler
    Absoft Corp Fortran Compiler
    MIPSpro Fortran Compiler
    Sun|Forte Fortran 95 Compiler
    Intel Fortran Compiler for 32-bit apps
    Intel Visual Fortran Compiler for 32-bit apps
    Intel Fortran Compiler for Itanium apps
    NAGWare Fortran 95 Compiler
    Compaq Fortran Compiler
    DIGITAL|Compaq Visual Fortran Compiler
    Pacific-Sierra Research Fortran 90 Compiler
    HP Fortran 90 Compiler
    Lahey/Fujitsu Fortran 95 Compiler
    IBM XL Fortran Compiler

* Numerous bugs are fixed. You should definitely update F2PY when using
  complex input arrays, there was a nasty bug that in certain cases
  caused incorrect results.

* F2PY has now a man page and its documentation is kept up to date. Many
  other improvements to F2PY algorithm and usage are implemented, see
  HISTORY.txt for more details.

Enjoy,
	Pearu Peterson

---------------
<P><A HREF="http://cens.ioc.ee/projects/f2py2e/">F2PY 2.39.235_1642</A> -
The Fortran to Python Interface Generator (24-Feb-04)



From oliphant at ee.byu.edu  Tue Feb 24 13:33:44 2004
From: oliphant at ee.byu.edu (Travis Oliphant)
Date: Tue, 24 Feb 2004 11:33:44 -0700
Subject: [SciPy-user] Help with filter design
In-Reply-To: <9B40DB5FBF7D2048AD79C2A91F862A5106DCFF@STEVAPPS05.Stevenage.CGI-Europe.com>
References: <9B40DB5FBF7D2048AD79C2A91F862A5106DCFF@STEVAPPS05.Stevenage.CGI-Europe.com>
Message-ID: <403B9908.50705@ee.byu.edu>

Bob.Cowdery at CGI-Europe.com wrote:

> Forgive these questions if they are rather basic. I am trying to get 
> up to speed with python and the scientific packages. I have an 
> implementation of a dsp process for a software radio implemented using 
> Intel IPP. I am converting this to python. One of the things I need to 
> do is fast convolution filtering. To create the filter in IPP I do 
> something like:
>
> ippsFIRGenLowpass_64f( ( high_cutoff - low_cutoff 
> )/2, taps, no_of_taps, window_type ) ) where taps is the output array.
>
> Is there something similar in SciPy or other packages. I found the 
> tutorial in the signal processing area too difficult to follow and I 
> can't seem to locate any other documentation for SciPy except the help 
> where you really need to know what you are looking for.
>
I would be interested to know which parts of the tutorial you found 
difficulty following and how you think it could be improved.

To answer your question...

Yes, there are quite a few filtering functions in SciPy.

There is signal.convolve which will convolve two functions together (an 
FIR filter)

There is also signal.lfilter which will implement an IIR filter (given 
the numerator and denominator polynomials)

It looks like FIRGenLowPass is creating an FIR filter.  Do you know what 
algorithm it is using?  What are the possibilities in window_type?

If you want to create an FIR filter then you should look at 
signal.remez  which will construct an FIR filter to minimize the maximum 
distance between your desired response in the pass band and the obtained 
response.   You can design low-pass, high-pass and bandpass filters 
using this method.

info(signal.remez) will tell you how to call it but it does assume some 
familiarity with designing digital filters.

You can also construct low-pass filters using the Fourier Transform and 
signal.get_window() 

signal.get_window() will construct a variety of windows for you.  You 
could use these windows as low-pass filters themselves or use them in 
the Fourier domain to construct an FIR filter (more common).


Let me know which approach you would prefer to use and I can give you 
more specific examples if you like.

-Travis Oliphant



From paxcalpt at sapo.pt  Wed Feb 25 02:58:38 2004
From: paxcalpt at sapo.pt (Ricardo Henriques)
Date: Wed, 25 Feb 2004 07:58:38 -0000
Subject: [SciPy-user] Does Scipy work well with py2exe?
Message-ID: <002501c3fb75$2f6a31e0$16b2c151@xcal>

Can anyone tell me if py2exe supports scipy importing?

Tks...

Paxcal


From Bob.Cowdery at CGI-Europe.com  Wed Feb 25 08:19:25 2004
From: Bob.Cowdery at CGI-Europe.com (Bob.Cowdery at CGI-Europe.com)
Date: Wed, 25 Feb 2004 13:19:25 -0000
Subject: [SciPy-user] FIR Filters
Message-ID: <9B40DB5FBF7D2048AD79C2A91F862A51772F66@STEVAPPS05.Stevenage.CGI-Europe.com>

I am trying to get a windowed FIR filter to use in fast convolution
filtering.
 
I have tried using :
signal.remez(2048,[300,3000],[1],Hz=44100)
to generate a 2048 tap filter, which is what I have used previously in Intel
IPP. Firstly, this gives me an array full of QNAN errors. Infact the only
time I don't get errors is by using a very small number of taps, say 10.
Various other values of taps fails to converge or gives QNAN errors. With
the 2048 taps it is also very, very slow (about 75 seconds on a 1.2GHz). The
IPP implementation is sub-second. I don't know if it's throwing an exception
on every QNAN or quite what is hapenning. 
 
I am also confused by not being able to specify a window type and I really
need lowpass not bandpass. I can see there is a function get_window() but I
don't know how this relates to an FIR filter when the only other parameter
is fftbins. Do I do something like tell it the number of fftbins I want the
filter to cover (I have 2048 bins of around 10Hz each) then shift it to the
correct part of the spectrum?
 
Any help appreciated.
 
Bob
 
 
Bob Cowdery
CGI Senior Technical Architect
+44(0)1438 791517
Mobile: +44(0)7771 532138
bob.cowdery at cgi-europe.com <mailto:bob.cowdery at cgi-europe.com> 
 
 
 

*** Confidentiality Notice *** Proprietary/Confidential
Information belonging to CGI Group Inc. and its affiliates
may be contained in this message. If you are not a recipient
indicated or intended in this message (or responsible for
delivery of this message to such person), or you think for
any reason that this message may have been addressed to you
in error, you may not use or copy or deliver this message
to anyone else.  In such case, you should destroy this
message and are asked to notify the sender by reply email.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.scipy.org/pipermail/scipy-user/attachments/20040225/ed30e5aa/attachment.html>

From arnd.baecker at web.de  Wed Feb 25 12:12:07 2004
From: arnd.baecker at web.de (Arnd Baecker)
Date: Wed, 25 Feb 2004 18:12:07 +0100 (CET)
Subject: [SciPy-user] scipy.xplt.ghelp (2nd try ;-)
Message-ID: <Pine.LNX.4.51.0402251807190.7538@ptpcp2.phy.tu-dresden.de>

Hi,

this is a reformulation of my recent question with the same subject,
I hope I get it right this time ;-)

If I understand things correctly, scipy.xplt.ghelp is gone
because the information from xplt/gistdata/help.help
has been moved to the doc strings of the commands.
If this is correct,
how do I get information on the keywords for a plot command ?

E.g.
  help("scipy.xplt.plg")
works, but I don't know how to get information on the keywords,
like `color`, `closed` etc.
I think all the information is given here
  http://bonsai.ims.u-tokyo.ac.jp/~mdehoon/software/python/pygist_html/node57.html
How can I access this type of information from within python?
(Before scipy.xplt.ghelp("color") did the job).

Many thanks,

Arnd


From gry at ll.mit.edu  Wed Feb 25 14:01:50 2004
From: gry at ll.mit.edu (george young)
Date: Wed, 25 Feb 2004 14:01:50 -0500
Subject: [SciPy-user] data smoothing: interpolate.splrep ignores s parameter
Message-ID: <20040225140150.0c97d070.gry@ll.mit.edu>

[SciPy-0.2.0_alpha_200.4161, Numeric-23.1, Python 2.3.3, x86 Linux]
My goal really is to smooth some noisy measurement data without messing
up it's *shape*.  My first attempt was 1d splines.  I did:

from Numeric import *
import scipy
f = file('rough')
data = array([(float(x),float(y)) for (x,y) in [l.split() for l in f]])
rep = scipy.interpolate.splrep(data[:,0], data[:,1], s=s)
smooth_y = scipy.interpolate.splev(data[:,0], rep)

"s" is supposed to vary the amount of smoothing.  For s=0, i get
my original data back, as expected.  But for all other values, including
the recommended range of (m-sqrt(2.0*m)) to (m+sqrt(2.0*m), I get a
single, too much smoothed, result.  It seems like splrep is not using
the "s" value to adjust the splines, except to sense that it's not zero.

Splines may not be the right method anyway, since they tend to warp the
shape of the curve, and I need to get the data's derivatives.  Is there
some way to fashion a low pass filter?  It seems like fft should be useful
here, but I have very little experience with fft's.

-- George Young

-- 
"Are the gods not just?"  "Oh no, child.
What would become of us if they were?" (CSL)


From gry at ll.mit.edu  Wed Feb 25 14:21:26 2004
From: gry at ll.mit.edu (george young)
Date: Wed, 25 Feb 2004 14:21:26 -0500
Subject: [SciPy-user] 
	Re: data smoothing: interpolate.splrep ignores s parameter
Message-ID: <20040225142126.096ed95e.gry@ll.mit.edu>

[[reposting with "rough" data file appended, sorry]]
[SciPy-0.2.0_alpha_200.4161, Numeric-23.1, Python 2.3.3, x86 Linux]
My goal really is to smooth some noisy measurement data without messing
up it's *shape*.  My first attempt was 1d splines.  I did:

from Numeric import *
import scipy
f = file('rough')
data = array([(float(x),float(y)) for (x,y) in [l.split() for l in f]])
rep = scipy.interpolate.splrep(data[:,0], data[:,1], s=s)
smooth_y = scipy.interpolate.splev(data[:,0], rep)

"s" is supposed to vary the amount of smoothing.  For s=0, i get
my original data back, as expected.  But for all other values, including
the recommended range of (m-sqrt(2.0*m)) to (m+sqrt(2.0*m), I get a
single, too much smoothed, result.  It seems like splrep is not using
the "s" value to adjust the splines, except to sense that it's not zero.

Splines may not be the right method anyway, since they tend to warp the
shape of the curve, and I need to get the data's derivatives.  Is there
some way to fashion a low pass filter?  It seems like fft should be useful
here, but I have very little experience with fft's.

-- George Young

"rough" file:
0.5 -1.7355e-10
0.455 -4.6185e-10
0.41 -3.2125e-10
0.365 -4.035e-10
0.32 2.951e-10
0.275 2.628e-10
0.23 2.922e-10
0.185 -3.1775e-10
0.14 -3.3209e-10
0.095 -3.6665e-10
0.05 -2.7394e-10
0.0 2.98e-10
-0.04 3.228e-10
-0.085 3.219e-10
-0.13 4.093e-10
-0.175 2.7406e-10
-0.22 1.102e-10
-0.265 -6.012e-10
-0.31 -3.9674e-09
-0.355 -1.2145e-08
-0.4 -3.9299e-08
-0.445 -1.2074e-07
-0.49 -3.2109e-07
-0.535 -7.3798e-07
-0.58 -1.4695e-06
-0.625 -2.548e-06
-0.67 -3.9104e-06
-0.715 -5.4895e-06
-0.76 -7.1758e-06
-0.805 -8.9267e-06
-0.85 -1.0675e-05
-0.895 -1.2425e-05
-0.94 -1.4115e-05
-0.985 -1.579e-05
-1.03 -1.7415e-05
-1.075 -1.9005e-05
-1.12 -2.0505e-05
-1.165 -2.2e-05
-1.21 -2.346e-05
-1.255 -2.4885e-05
-1.3 -2.624e-05
-1.345 -2.757e-05
-1.39 -2.886e-05
-1.435 -3.015e-05
-1.48 -3.1404e-05
-1.525 -3.2604e-05
-1.57 -3.385e-05
-1.615 -3.498e-05
-1.66 -3.6105e-05
-1.705 -3.718e-05
-1.75 -3.8279e-05
-1.795 -3.932e-05
-1.84 -4.033e-05
-1.885 -4.13e-05
-1.93 -4.2304e-05
-1.975 -4.3275e-05
-2.02 -4.429e-05
-2.065 -4.5205e-05
-2.11 -4.6069e-05
-2.155 -4.697e-05
-2.2 -4.7825e-05
-2.245 -4.8639e-05
-2.29 -4.9515e-05
-2.335 -5.0319e-05
-2.38 -5.1094e-05
-2.425 -5.188e-05
-2.47 -5.264e-05
-2.515 -5.3454e-05
-2.56 -5.4169e-05
-2.605 -5.4905e-05
-2.65 -5.5633e-05
-2.695 -5.6334e-05
-2.74 -5.7043e-05
-2.785 -5.7749e-05
-2.83 -5.8343e-05
-2.875 -5.9085e-05
-2.92 -5.9729e-05
-2.965 -6.0344e-05
-3.01 -6.094e-05
-3.055 -6.1564e-05
-3.1 -6.2104e-05
-3.145 -6.2749e-05
-3.19 -6.3348e-05
-3.235 -6.3956e-05
-3.28 -6.4414e-05
-3.325 -6.4958e-05
-3.37 -6.555e-05
-3.415 -6.6075e-05
-3.46 -6.656e-05
-3.505 -6.71e-05
-3.55 -6.7603e-05
-3.595 -6.8083e-05
-3.64 -6.8623e-05
-3.685 -6.9048e-05
-3.73 -6.9518e-05
-3.775 -6.998e-05
-3.82 -7.0508e-05
-3.865 -7.0929e-05
-3.91 -7.1343e-05
-3.955 -7.1824e-05
-4.0 -7.2218e-05


From pearu at scipy.org  Wed Feb 25 15:45:53 2004
From: pearu at scipy.org (Pearu Peterson)
Date: Wed, 25 Feb 2004 14:45:53 -0600 (CST)
Subject: [SciPy-user]  Re: data smoothing: interpolate.splrep ignores s
 parameter
In-Reply-To: <20040225142126.096ed95e.gry@ll.mit.edu>
Message-ID: <Pine.LNX.4.44.0402251431480.31709-100000@scipy.org>



On Wed, 25 Feb 2004, george young wrote:

> [[reposting with "rough" data file appended, sorry]]
> [SciPy-0.2.0_alpha_200.4161, Numeric-23.1, Python 2.3.3, x86 Linux]

I recommend using CVS version of Scipy.

> My goal really is to smooth some noisy measurement data without messing
> up it's *shape*.  My first attempt was 1d splines.  I did:
> 
> from Numeric import *
> import scipy
> f = file('rough')
> data = array([(float(x),float(y)) for (x,y) in [l.split() for l in f]])
> rep = scipy.interpolate.splrep(data[:,0], data[:,1], s=s)
> smooth_y = scipy.interpolate.splev(data[:,0], rep)
> 
> "s" is supposed to vary the amount of smoothing.  For s=0, i get
> my original data back, as expected.  But for all other values, including
> the recommended range of (m-sqrt(2.0*m)) to (m+sqrt(2.0*m), I get a
> single, too much smoothed, result.  It seems like splrep is not using
> the "s" value to adjust the splines, except to sense that it's not zero.

First note that you have to specify x data as an increasing sequence of 
numbers, currently data[:,0] is decreasing.

Second, your data is quite smooth, so you must use very small s values to 
see the effect, e.g. try for example s=1e-11 or smaller.

> Splines may not be the right method anyway, since they tend to warp the
> shape of the curve, and I need to get the data's derivatives.  Is there
> some way to fashion a low pass filter?  It seems like fft should be useful
> here, but I have very little experience with fft's.

FFT is not suitable here as your data is not periodic. But if you need 
only derivatives, then use

rep = \
 interpolate.fitpack2.UnivariateSpline(data[::-1,0],data[::-1,1],s=1e-13)

y_der = rep(data[:,0],1) # first derivative

You may need to play with s value a bit and may be also split your data 
around x=-0.5 and do the smoothing on these parts separately because the 
data behaves differently on those two parts.

HTH,
Pearu


From pearu at scipy.org  Wed Feb 25 16:09:55 2004
From: pearu at scipy.org (Pearu Peterson)
Date: Wed, 25 Feb 2004 15:09:55 -0600 (CST)
Subject: [SciPy-user] scipy.xplt.ghelp (2nd try ;-)
In-Reply-To: <Pine.LNX.4.51.0402251807190.7538@ptpcp2.phy.tu-dresden.de>
Message-ID: <Pine.LNX.4.44.0402251507530.31709-100000@scipy.org>



On Wed, 25 Feb 2004, Arnd Baecker wrote:

> Hi,
> 
> this is a reformulation of my recent question with the same subject,
> I hope I get it right this time ;-)
> 
> If I understand things correctly, scipy.xplt.ghelp is gone
> because the information from xplt/gistdata/help.help
> has been moved to the doc strings of the commands.
> If this is correct,
> how do I get information on the keywords for a plot command ?
> 
> E.g.
>   help("scipy.xplt.plg")
> works, but I don't know how to get information on the keywords,
> like `color`, `closed` etc.
> I think all the information is given here
>   http://bonsai.ims.u-tokyo.ac.jp/~mdehoon/software/python/pygist_html/node57.html
> How can I access this type of information from within python?
> (Before scipy.xplt.ghelp("color") did the job).

I am not familiar with the background of loosing scipy.xplt.ghelp but the 
following works:

import scipy.xplt.helpmod
scipy.xplt.helpmod.help('color')

Pearu


From oliphant at ee.byu.edu  Wed Feb 25 17:29:44 2004
From: oliphant at ee.byu.edu (Travis E. Oliphant)
Date: Wed, 25 Feb 2004 15:29:44 -0700
Subject: [SciPy-user] scipy.xplt.ghelp (2nd try ;-)
In-Reply-To: <Pine.LNX.4.51.0402251807190.7538@ptpcp2.phy.tu-dresden.de>
References: <Pine.LNX.4.51.0402251807190.7538@ptpcp2.phy.tu-dresden.de>
Message-ID: <403D21D8.1000500@ee.byu.edu>

Arnd Baecker wrote:
> Hi,
> 
> this is a reformulation of my recent question with the same subject,
> I hope I get it right this time ;-)
> 
> If I understand things correctly, scipy.xplt.ghelp is gone
> because the information from xplt/gistdata/help.help
> has been moved to the doc strings of the commands.
> If this is correct,

Yes, that was the idea.

> how do I get information on the keywords for a plot command ?

It looks like we have a problem, as the docstrings aren't complete yet.

I should add back the ghelp command in the meantime.

You can get it from an older version of the code. I will try to update 
the CVS (but it may be a week or two).

Thanks for the heads-up.

-Travis


From val at vtek.com  Wed Feb 25 17:47:20 2004
From: val at vtek.com (val)
Date: Wed, 25 Feb 2004 17:47:20 -0500
Subject: [SciPy-user] Re: data smoothing: interpolate.splrep ignores s
	parameter
References: <20040225142126.096ed95e.gry@ll.mit.edu>
Message-ID: <070f01c3fbf1$54654940$6400a8c0@vt1000>

Well,
    Maybe, first you'd set up the realistic precision
eps you want.  For now, you have a range of 5 orders
of magnitudes.  Scale your data based on eps and graph
it to see how noisy the data is.  For physical variables,
the noise should be within reasonable range.  Smoothing
'per se' does not help much with unreasonably noisy data.
    Or i'm missing smth?
val


----- Original Message -----
From: "george young" <gry at ll.mit.edu>
To: <scipy-user at scipy.net>
Sent: Wednesday, February 25, 2004 2:21 PM
Subject: [SciPy-user] Re: data smoothing: interpolate.splrep ignores s
parameter


> [[reposting with "rough" data file appended, sorry]]
> [SciPy-0.2.0_alpha_200.4161, Numeric-23.1, Python 2.3.3, x86 Linux]
> My goal really is to smooth some noisy measurement data without messing
> up it's *shape*.  My first attempt was 1d splines.  I did:
>
> from Numeric import *
> import scipy
> f = file('rough')
> data = array([(float(x),float(y)) for (x,y) in [l.split() for l in f]])
> rep = scipy.interpolate.splrep(data[:,0], data[:,1], s=s)
> smooth_y = scipy.interpolate.splev(data[:,0], rep)
>
> "s" is supposed to vary the amount of smoothing.  For s=0, i get
> my original data back, as expected.  But for all other values, including
> the recommended range of (m-sqrt(2.0*m)) to (m+sqrt(2.0*m), I get a
> single, too much smoothed, result.  It seems like splrep is not using
> the "s" value to adjust the splines, except to sense that it's not zero.
>
> Splines may not be the right method anyway, since they tend to warp the
> shape of the curve, and I need to get the data's derivatives.  Is there
> some way to fashion a low pass filter?  It seems like fft should be useful
> here, but I have very little experience with fft's.
>
> -- George Young
>
> "rough" file:
> 0.5 -1.7355e-10
> 0.455 -4.6185e-10
> 0.41 -3.2125e-10
> 0.365 -4.035e-10
> 0.32 2.951e-10
> 0.275 2.628e-10
> 0.23 2.922e-10
> 0.185 -3.1775e-10
> 0.14 -3.3209e-10
> 0.095 -3.6665e-10
> 0.05 -2.7394e-10
> 0.0 2.98e-10
> -0.04 3.228e-10
> -0.085 3.219e-10
> -0.13 4.093e-10
> -0.175 2.7406e-10
> -0.22 1.102e-10
> -0.265 -6.012e-10
> -0.31 -3.9674e-09
> -0.355 -1.2145e-08
> -0.4 -3.9299e-08
> -0.445 -1.2074e-07
> -0.49 -3.2109e-07
> -0.535 -7.3798e-07
> -0.58 -1.4695e-06
> -0.625 -2.548e-06
> -0.67 -3.9104e-06
> -0.715 -5.4895e-06
> -0.76 -7.1758e-06
> -0.805 -8.9267e-06
> -0.85 -1.0675e-05
> -0.895 -1.2425e-05
> -0.94 -1.4115e-05
> -0.985 -1.579e-05
> -1.03 -1.7415e-05
> -1.075 -1.9005e-05
> -1.12 -2.0505e-05
> -1.165 -2.2e-05
> -1.21 -2.346e-05
> -1.255 -2.4885e-05
> -1.3 -2.624e-05
> -1.345 -2.757e-05
> -1.39 -2.886e-05
> -1.435 -3.015e-05
> -1.48 -3.1404e-05
> -1.525 -3.2604e-05
> -1.57 -3.385e-05
> -1.615 -3.498e-05
> -1.66 -3.6105e-05
> -1.705 -3.718e-05
> -1.75 -3.8279e-05
> -1.795 -3.932e-05
> -1.84 -4.033e-05
> -1.885 -4.13e-05
> -1.93 -4.2304e-05
> -1.975 -4.3275e-05
> -2.02 -4.429e-05
> -2.065 -4.5205e-05
> -2.11 -4.6069e-05
> -2.155 -4.697e-05
> -2.2 -4.7825e-05
> -2.245 -4.8639e-05
> -2.29 -4.9515e-05
> -2.335 -5.0319e-05
> -2.38 -5.1094e-05
> -2.425 -5.188e-05
> -2.47 -5.264e-05
> -2.515 -5.3454e-05
> -2.56 -5.4169e-05
> -2.605 -5.4905e-05
> -2.65 -5.5633e-05
> -2.695 -5.6334e-05
> -2.74 -5.7043e-05
> -2.785 -5.7749e-05
> -2.83 -5.8343e-05
> -2.875 -5.9085e-05
> -2.92 -5.9729e-05
> -2.965 -6.0344e-05
> -3.01 -6.094e-05
> -3.055 -6.1564e-05
> -3.1 -6.2104e-05
> -3.145 -6.2749e-05
> -3.19 -6.3348e-05
> -3.235 -6.3956e-05
> -3.28 -6.4414e-05
> -3.325 -6.4958e-05
> -3.37 -6.555e-05
> -3.415 -6.6075e-05
> -3.46 -6.656e-05
> -3.505 -6.71e-05
> -3.55 -6.7603e-05
> -3.595 -6.8083e-05
> -3.64 -6.8623e-05
> -3.685 -6.9048e-05
> -3.73 -6.9518e-05
> -3.775 -6.998e-05
> -3.82 -7.0508e-05
> -3.865 -7.0929e-05
> -3.91 -7.1343e-05
> -3.955 -7.1824e-05
> -4.0 -7.2218e-05
>
> _______________________________________________
> SciPy-user mailing list
> SciPy-user at scipy.net
> http://www.scipy.net/mailman/listinfo/scipy-user
>


From oliphant at ee.byu.edu  Wed Feb 25 18:40:34 2004
From: oliphant at ee.byu.edu (Travis E. Oliphant)
Date: Wed, 25 Feb 2004 16:40:34 -0700
Subject: [SciPy-user] FIR Filters
In-Reply-To: <9B40DB5FBF7D2048AD79C2A91F862A51772F66@STEVAPPS05.Stevenage.CGI-Europe.com>
References: <9B40DB5FBF7D2048AD79C2A91F862A51772F66@STEVAPPS05.Stevenage.CGI-Europe.com>
Message-ID: <403D3272.5050009@ee.byu.edu>

Bob.Cowdery at CGI-Europe.com wrote:
> I am trying to get a windowed FIR filter to use in fast convolution 
> filtering.
>  
> I have tried using :
> signal.remez(2048,[300,3000],[1],Hz=44100)
> to generate a 2048 tap filter, which is what I have used previously in 
> Intel IPP. Firstly, this gives me an array full of QNAN errors. 

That is a lot of taps.  In theory remez should work, but I admit it has 
not been tested for so many taps.   And, the "bandpass" option can 
generate a lowpass filter by making the lower band 0.

Why are you doing convolution filtering with 2048 taps?   Why are you 
not using the FFT which should be much faster?

> Infact 
> the only time I don't get errors is by using a very small number of 
> taps, say 10. Various other values of taps fails to converge or gives 
> QNAN errors. With the 2048 taps it is also very, very slow (about 75 
> seconds on a 1.2GHz). The IPP implementation is sub-second. I don't know 
> if it's throwing an exception on every QNAN or quite what is hapenning.

I don't think IPP is using the remez algorithm for getting it's filter. 
    It's probably doing windowed filter design.

> I am also confused by not being able to specify a window type and I 
> really need lowpass not bandpass. I can see there is a function 
> get_window() but I don't know how this relates to an FIR filter when the 
> only other parameter is fftbins. Do I do something like tell it the 
> number of fftbins I want the filter to cover (I have 2048 bins of around 
> 10Hz each) then shift it to the correct part of the spectrum?


There is not currently an FIR-filter design program in SciPy.  One 
should be constructed as it is not hard to implement (of course making 
it generic with all the options you might want would take some time).

What kind of window are you currently using?

For your purposes this is what I would do:

w = fftfreq(2048,T)  # give frequency bins in Hz, T is sample spacing

myfilter = where((abs(w)< cutoff),1,0)  # cutoff is low-pass filter

h = ifft(myfilter) # ideal filter

beta = 11.7
myh = fftshift(h) * get_window(beta,2048)  # beta implies Kaiser Filter


Someday this will get wrapped up into a nice FIR filter design routine 
--- any suggestions on the interface that routine should take would be 
useful.

-Travis O.









>  
> Any help appreciated.
>  
> Bob
>  
>  
> Bob Cowdery
> CGI Senior Technical Architect
> +44(0)1438 791517
> Mobile: +44(0)7771 532138
> bob.cowdery at cgi-europe.com <mailto:bob.cowdery at cgi-europe.com>
>  
>  
>  
> 
> **** Confidentiality Notice **** Proprietary/Confidential
> Information belonging to CGI Group Inc. and its affiliates
> may be contained in this message. If you are not a recipient
> indicated or intended in this message (or responsible for
> delivery of this message to such person), or you think for
> any reason that this message may have been addressed to you
> in error, you may not use or copy or deliver this message
> to anyone else.  In such case, you should destroy this
> message and are asked to notify the sender by reply email.
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> SciPy-user mailing list
> SciPy-user at scipy.net
> http://www.scipy.net/mailman/listinfo/scipy-user


From oliphant at ee.byu.edu  Wed Feb 25 18:43:30 2004
From: oliphant at ee.byu.edu (Travis E. Oliphant)
Date: Wed, 25 Feb 2004 16:43:30 -0700
Subject: [SciPy-user]  Re: data smoothing: interpolate.splrep ignores
 s parameter
In-Reply-To: <Pine.LNX.4.44.0402251431480.31709-100000@scipy.org>
References: <Pine.LNX.4.44.0402251431480.31709-100000@scipy.org>
Message-ID: <403D3322.9080304@ee.byu.edu>

Pearu Peterson wrote:
> 
> On Wed, 25 Feb 2004, george young wrote:
> 
> FFT is not suitable here as your data is not periodic. But if you need 
> only derivatives, then use

This is not necessarily true.  The FFT can be used for non-periodic data 
  -- and is in fact used quite often for such cases as long as you are 
careful.  You just have to realize that the FFT is assuming your data is 
periodic and so any filtering (i.e. smoothing) will be using wrap-around 
boundary conditions.

You can window your data, zero-pad, mirror it on the boundaries and many 
other things in order to get rid of artifacts that result from the 
periodic assumption of the FFT.


-Travis O.


From jbenson at sextans.lowell.edu  Wed Feb 25 22:53:26 2004
From: jbenson at sextans.lowell.edu (Jim Benson)
Date: Wed, 25 Feb 2004 20:53:26 -0700 (MST)
Subject: [SciPy-user] scipy cvs snapshot problem
Message-ID: <Pine.LNX.4.44.0402252031070.4666-100000@sextans.mesa.lowellnet>


Hi,

I downloaded a cvs snapshot on 2004-02-25 and installed it
over a previous cvs snapshot (from 2003-12-03). The previous 
one did pass all scipy.test levels. The recent version:

Found scipy_distutils version '0.2.2_alpha_26.374' in 
'/usr/local/lib/python2.3/
site-packages/scipy_distutils/__init__.pyc'
Importing scipy_distutils.command.build_flib ... ok

failed the first test:

======================================================================
FAIL: check_nils_20Feb04 (scipy.linalg.basic.test_basic.test_solve)
----------------------------------------------------------------------
Traceback (most recent call last):
  File 
"/usr/local/lib/python2.3/site-packages/scipy/linalg/tests/test_basic.py", 
line 112, in check_nils_20Feb04
    assert_array_almost_equal(X,Ainv)
  File "/usr/local/lib/python2.3/site-packages/scipy_test/testing.py", 
line 636, in assert_array_almost_equal
    assert cond,\
AssertionError:
Arrays are not almost equal (mismatch 50.0%):
        Array 1: [[-2.3261187+2.4103813j  2.6175944-0.5570154j]
 [ 2.9991264-1.7766662j -2.0232508-0.2587063j]]
        Array 2: [[-2.8831341-0.2072131j  2.6175944-0.5570154j]
 [ 2.7404201+0.2465847j -2.0232508-0.2587063j]]
 
 
----------------------------------------------------------------------
Ran 1198 tests in 1.918s
 
FAILED (failures=1)
<unittest.TextTestRunner object at 0x4240ef2c>
>>>

My atlas version: 

floyd:/usr/local/scipy/Lib/linalg>python -c 'import atlas_version'
ATLAS version 3.4.2 built by root on Tue Dec  2 20:35:31 MST 2003:
 

...am i forgetting something critical?

Thanks,
 
Jim



From oliphant at ee.byu.edu  Thu Feb 26 02:09:56 2004
From: oliphant at ee.byu.edu (Travis Oliphant)
Date: Thu, 26 Feb 2004 00:09:56 -0700
Subject: [SciPy-user] FIR Filters
In-Reply-To: <9B40DB5FBF7D2048AD79C2A91F862A51772F66@STEVAPPS05.Stevenage.CGI-Europe.com>
References: <9B40DB5FBF7D2048AD79C2A91F862A51772F66@STEVAPPS05.Stevenage.CGI-Europe.com>
Message-ID: <403D9BC4.8000808@ee.byu.edu>


> I am trying to get a windowed FIR filter to use in fast convolution 
> filtering.
>  
> I have tried using :
> signal.remez(2048,[300,3000],[1],Hz=44100)
> to generate a 2048 tap filter, which is what I have used previously in 
> Intel IPP. Firstly, this gives me an array full of QNAN errors. Infact 
> the only time I don't get errors is by using a very small number of 
> taps, say 10. Various other values of taps fails to converge or gives 
> QNAN errors. With the 2048 taps it is also very, very slow (about 75 
> seconds on a 1.2GHz). The IPP implementation is sub-second. I don't 
> know if it's throwing an exception on every QNAN or quite what is 
> hapenning.
>  
> I am also confused by not being able to specify a window type and I 
> really need lowpass not bandpass. I can see there is a function 
> get_window() but I don't know how this relates to an FIR filter when 
> the only other parameter is fftbins. Do I do something like tell it 
> the number of fftbins I want the filter to cover (I have 2048 bins of 
> around 10Hz each) then shift it to the correct part of the spectrum?


Bob,

Here is a simple windowed ideal lowpass filter algorithm.  The two 
important parameters are the number of taps and the cutoff frequency 
normalized
  so that 1 corresponds to pi radians/sample (i.e. the Nyquist 
frequency).  If you specify width it will choose a nice Kaiser window 
for you to try and reach
  that width of transistion region (from  passband to stopband).  
Otherwise, you can specify the window type (see signal.get_window for types)

This is now in SciPy CVS.



def firwin(N, cutoff, width=None, window='hamming'):
    """FIR Filter Design using windowed ideal filter method.

    Inputs:
   
      N      -- order of filter (number of taps)
      cutoff -- cutoff frequency of filter (normalized so that 1 
corresponds to
                  Nyquist or pi radians / sample)

      width  -- if width is not None, then assume it is the approximate 
width of
                  the transition region (normalized so that 1 corresonds 
to pi)
                  for use in kaiser FIR filter design.
      window -- desired window to use.

    Outputs:

      h      -- coefficients of length N fir filter.
    """

    from signaltools import get_window
    if isinstance(width,float):
        A = 2.285*N*width + 8
        if (A < 21): beta = 0.0
        elif (A <= 50): beta = 0.5842*(A-21)**0.4 + 0.07886*(A-21)
        else: beta = 0.1102*(A-8.7)
        window=('kaiser',beta)

    win = get_window(window,N,fftbins=1)
    alpha = N//2
    m = Num.arange(0,N)
    return win*special.sinc(cutoff*(m-alpha))
   
   


From pearu at scipy.org  Thu Feb 26 03:52:26 2004
From: pearu at scipy.org (Pearu Peterson)
Date: Thu, 26 Feb 2004 02:52:26 -0600 (CST)
Subject: [SciPy-user] scipy cvs snapshot problem
In-Reply-To: <Pine.LNX.4.44.0402252031070.4666-100000@sextans.mesa.lowellnet>
Message-ID: <Pine.LNX.4.44.0402260233520.31709-100000@scipy.org>


On Wed, 25 Feb 2004, Jim Benson wrote:

> I downloaded a cvs snapshot on 2004-02-25 and installed it
> over a previous cvs snapshot (from 2003-12-03). The previous 
> one did pass all scipy.test levels. The recent version
> failed the first test:
> 
> ======================================================================
> FAIL: check_nils_20Feb04 (scipy.linalg.basic.test_basic.test_solve)
> ----------------------------------------------------------------------
> AssertionError:
> Arrays are not almost equal (mismatch 50.0%):
>         Array 1: [[-2.3261187+2.4103813j  2.6175944-0.5570154j]
>  [ 2.9991264-1.7766662j -2.0232508-0.2587063j]]
>         Array 2: [[-2.8831341-0.2072131j  2.6175944-0.5570154j]
>  [ 2.7404201+0.2465847j -2.0232508-0.2587063j]]

> ...am i forgetting something critical?

Yes, you missed recent thread about the corner case where linalg.solve 
could return wrong results. The above test covers this case.
You should certainly update f2py to the latest if somewhere in your 
applications the following conditions may be holding:

  a complex one-dimensional slice of an array is passed to f2py
  generated wrapper.

Actually as far as I am concerned, the origin of this bug lies in 
all versions of Numeric CDOUBLE_to_CDOUBLE,etc functions for
any(!) non-contiguous complex array.
 
This bug was reported to numpy list years ago, see

  http://sourceforge.net/mailarchive/message.php?msg_id=1282411

but since there was no response or interest to it :(, f2py provided a 
workaround to this problem but missed the above important corner
case.

Regards,
Pearu


From arnd.baecker at web.de  Thu Feb 26 05:34:34 2004
From: arnd.baecker at web.de (Arnd Baecker)
Date: Thu, 26 Feb 2004 11:34:34 +0100 (CET)
Subject: [SciPy-user] scipy.xplt.ghelp (2nd try ;-)
In-Reply-To: <Pine.LNX.4.44.0402251507530.31709-100000@scipy.org>
References: <Pine.LNX.4.44.0402251507530.31709-100000@scipy.org>
Message-ID: <Pine.LNX.4.51.0402261130240.20086@ptpcp2.phy.tu-dresden.de>


On Wed, 25 Feb 2004, Pearu Peterson wrote:
[...]

> > How can I access this type of information from within python?
> > (Before scipy.xplt.ghelp("color") did the job).
>
> I am not familiar with the background of loosing scipy.xplt.ghelp but the
> following works:
>
> import scipy.xplt.helpmod
> scipy.xplt.helpmod.help('color')

Thanks, Pearu! I also just checked that putting a
  from helpmod import help as ghelp
into gist.py
restores the original behaviour, i.e. scipy.xplt.ghelp("color")
works as before.
So this is an easy temporary fix, but I think now that
all the doc-strings are there for the commands we
could with little effort also get the keywords into
one of the doc-strings (see my next message)

Best,

Arnd


From arnd.baecker at web.de  Thu Feb 26 05:57:03 2004
From: arnd.baecker at web.de (Arnd Baecker)
Date: Thu, 26 Feb 2004 11:57:03 +0100 (CET)
Subject: [SciPy-user] scipy.xplt.ghelp (2nd try ;-)
In-Reply-To: <403D21D8.1000500@ee.byu.edu>
References: <Pine.LNX.4.51.0402251807190.7538@ptpcp2.phy.tu-dresden.de>
 <403D21D8.1000500@ee.byu.edu>
Message-ID: <Pine.LNX.4.51.0402260125030.1088@ptplp09.phy.tu-dresden.de>


On Wed, 25 Feb 2004, Travis E. Oliphant wrote:

> Arnd Baecker wrote:
[...]
> > If I understand things correctly, scipy.xplt.ghelp is gone
> > because the information from xplt/gistdata/help.help
> > has been moved to the doc strings of the commands.
> > If this is correct,
>
> Yes, that was the idea.

I am very much in favour of this!

> > how do I get information on the keywords for a plot command ?
>
> It looks like we have a problem, as the docstrings aren't complete yet.

As far as I can see it is really only the keywords.
Thinking further about it, there might be
a  simple solution to this
  a) add the information for the different keywords to
     the docstring of scipy.xplt itself.
     One could then add the text
      "see help("scipy.xplt") for information on the keywords"
     in the doc-strings for the corresponding plot routines.
  b) add the information for the different keywords to
     the docstring of scipy.xplt.plg.
     One could then add the text
      "see help("scipy.xplt.plg") for information on the keywords"
     in the doc-strings for the corresponding other plot routines.
  c) Invent a dummy routine
       scipy.xplt.keywords
     which contains the information on the keywords.

Personally I don't like c) very much, so maybe b) is
the best solution as the doc-string will be pretty long
(i.e. a reason not to put it into scipy.xplt).
If you think that this is a reasonable (also long-term) solution
then I can prepare a corresponding doc-string.

One more general question on the doc-strings of xplt:
Some of the doc-strings (e.g. get_style(...) and many more)
are so long that for 80 characters display width
the corresponding lines get wrapped. To be precise:
the doc-strings themselves are (mostly) <80 chars,
but the indent when using the help brings the whole line to > 80 chars.
If you think this is worth doing, I could take
care of changing this in gist.py (and presuambly some other files).

> I should add back the ghelp command in the meantime.
>
> You can get it from an older version of the code. I will try to update
> the CVS (but it may be a week or two).

That's perfectly fine with me - many thanks,

Arnd


From Bob.Cowdery at CGI-Europe.com  Thu Feb 26 08:27:40 2004
From: Bob.Cowdery at CGI-Europe.com (Bob.Cowdery at CGI-Europe.com)
Date: Thu, 26 Feb 2004 13:27:40 -0000
Subject: [SciPy-user] FIR Filters
Message-ID: <9B40DB5FBF7D2048AD79C2A91F862A51772F68@STEVAPPS05.Stevenage.CGI-Europe.com>

Travis

Thanks very much for the responses. Don't confuse me with someone who knows
what they are doing (but I am working on it!). To give you some background I
am converting a .NET application that uses Intel IPP to Python. The
application is for a software defined radio. At the moment I have audio
capture which contains my quadrature samples in blocks of 4096 (2048 each
for I&Q) with 44100 sampling rate (46ms per block). I have a simple dsp
process to test each signal processing call. At the moment this should
output exactly what it receives, i.e just the audio input as the demodulator
needs the FIR filter and some magic applied at the point indicated in the
code. When I have written the demodulator I can test firwin() properly. At
the moment I can see it outputs valid data, and it's fast which is good. The
code below is the very simple path. I have had to write two functions for
polar/cartesian conversions as I didn't find any in the library. I don't
know if these work right as I can hear clicking sounds on the audio not
there when they are commented out. However, it could be audio overrun as I
am up at 75% processor. The functions are also quite slow (adding 30% to my
processor loading). Is there a more efficient way to do this?

    def sigproc( self, samples ) :        
        
        # convert to a numarray array of type double
        x = asarray(samples, typecode=Float64)
        
        # first to do is sort into real and image arrays from interleaved
array
        realin = x[1::2]  # every other element starting at index 1
        imagin = x[::2]   # every other element starting at index 0
                
        # convert to frequency domain
        realfd = fft(realin)
        imagfd = fft(imagin)
        
        # cartesian to polar in frequency domain
        magn,phase = self.cartesian_to_polar(realfd, imagfd)
        
        # apply filter and demodulate
                
        # polar to cartesian in frequency domain
        realfd, imagfd = self.polar_to_cartesian(magn, phase)
        
        # convert to time domain
        realtd = ifft(realfd)
        imagtd = ifft(imagfd)
                
        # convert back to interleaved array
        x[::2] = realtd.astype(Float64)
        x[1::2] = imagtd.astype(Float64)
        
        # convert back to an integer array
        x = x.astype('l')
        
        # return python list of processed samples
        return x.tolist()
    
    # utility functions not in the library
    def polar_to_cartesian(self, magn, phase):
        return magn * cos(phase), magn * sin(phase)

    def cartesian_to_polar(self, real, imag):
        # convert the arrays real, imag into the polar system
        # magnitude array
        magn = sqrt(real**2 + imag**2)
        # phase array
        phase =
choose(equal(real,0),(arctan(imag/real),(choose(less(imag,0),((pi/2),-(pi/2)
)))))
        return magn, phase


-----Original Message-----
From: Travis Oliphant [mailto:oliphant at ee.byu.edu] 
Sent: 26 February 2004 07:10
To: SciPy Users List
Subject: Re: [SciPy-user] FIR Filters



> I am trying to get a windowed FIR filter to use in fast convolution
> filtering.
>  
> I have tried using :
> signal.remez(2048,[300,3000],[1],Hz=44100)
> to generate a 2048 tap filter, which is what I have used previously in
> Intel IPP. Firstly, this gives me an array full of QNAN errors. Infact 
> the only time I don't get errors is by using a very small number of 
> taps, say 10. Various other values of taps fails to converge or gives 
> QNAN errors. With the 2048 taps it is also very, very slow (about 75 
> seconds on a 1.2GHz). The IPP implementation is sub-second. I don't 
> know if it's throwing an exception on every QNAN or quite what is 
> hapenning.
>  
> I am also confused by not being able to specify a window type and I
> really need lowpass not bandpass. I can see there is a function 
> get_window() but I don't know how this relates to an FIR filter when 
> the only other parameter is fftbins. Do I do something like tell it 
> the number of fftbins I want the filter to cover (I have 2048 bins of 
> around 10Hz each) then shift it to the correct part of the spectrum?


Bob,

Here is a simple windowed ideal lowpass filter algorithm.  The two 
important parameters are the number of taps and the cutoff frequency 
normalized
  so that 1 corresponds to pi radians/sample (i.e. the Nyquist 
frequency).  If you specify width it will choose a nice Kaiser window 
for you to try and reach
  that width of transistion region (from  passband to stopband).  
Otherwise, you can specify the window type (see signal.get_window for types)

This is now in SciPy CVS.



def firwin(N, cutoff, width=None, window='hamming'):
    """FIR Filter Design using windowed ideal filter method.

    Inputs:
   
      N      -- order of filter (number of taps)
      cutoff -- cutoff frequency of filter (normalized so that 1 
corresponds to
                  Nyquist or pi radians / sample)

      width  -- if width is not None, then assume it is the approximate 
width of
                  the transition region (normalized so that 1 corresonds 
to pi)
                  for use in kaiser FIR filter design.
      window -- desired window to use.

    Outputs:

      h      -- coefficients of length N fir filter.
    """

    from signaltools import get_window
    if isinstance(width,float):
        A = 2.285*N*width + 8
        if (A < 21): beta = 0.0
        elif (A <= 50): beta = 0.5842*(A-21)**0.4 + 0.07886*(A-21)
        else: beta = 0.1102*(A-8.7)
        window=('kaiser',beta)

    win = get_window(window,N,fftbins=1)
    alpha = N//2
    m = Num.arange(0,N)
    return win*special.sinc(cutoff*(m-alpha))
   
   

_______________________________________________
SciPy-user mailing list
SciPy-user at scipy.net
http://www.scipy.net/mailman/listinfo/scipy-user

*** Confidentiality Notice *** Proprietary/Confidential
Information belonging to CGI Group Inc. and its affiliates
may be contained in this message. If you are not a recipient
indicated or intended in this message (or responsible for
delivery of this message to such person), or you think for
any reason that this message may have been addressed to you
in error, you may not use or copy or deliver this message
to anyone else.  In such case, you should destroy this
message and are asked to notify the sender by reply email.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.scipy.org/pipermail/scipy-user/attachments/20040226/c79bdfc0/attachment.html>

From magnew at math.uio.no  Thu Feb 26 09:12:20 2004
From: magnew at math.uio.no (Magne Westlie)
Date: Thu, 26 Feb 2004 15:12:20 +0100 (CET)
Subject: [SciPy-user] weave, including headers in generated file
In-Reply-To: <mailman.0.1076943290.357.scipy-user@scipy.net>
References: <mailman.0.1076943290.357.scipy-user@scipy.net>
Message-ID: <Pine.LNX.4.58.0402261450130.24066@nelja.ifi.uio.no>


Hello.

I'm having a problem with using weave which I was wondering if someone
could help me out with.

I've got a class Hello declared in a file hw.h, but how do I tell weave to
include this file in the generated cpp-file? If I try:

code = """
Hello h;
h.hw();
"""

weave.inline(code,
             [],
             compiler='gcc',
             verbose=2,
             force=1,
             sources=['/home/mw/weave/inline-ex/hw.cpp'],
             include_dirs=['/home/mw/weave/inline-ex/'],
             )

Compiling fails because Hello is not defined:

/home/mw/.python23_compiled/sc_dbdc99b5e1f621cd206ee575625e18f15.cpp:644:
error: `
   Hello' undeclared (first use this function)


If I open the cpp file manually and puts:

#include <hw.h>

in it, everything works.



Magne Westlie


From barrett at stsci.edu  Thu Feb 26 09:15:58 2004
From: barrett at stsci.edu (Paul Barrett)
Date: Thu, 26 Feb 2004 09:15:58 -0500
Subject: [SciPy-user] 	Re: data smoothing: interpolate.splrep ignores
	s parameter
In-Reply-To: <20040225142126.096ed95e.gry@ll.mit.edu>
References: <20040225142126.096ed95e.gry@ll.mit.edu>
Message-ID: <403DFF9E.20403@stsci.edu>

george young wrote:
> [[reposting with "rough" data file appended, sorry]]
> [SciPy-0.2.0_alpha_200.4161, Numeric-23.1, Python 2.3.3, x86 Linux]
> My goal really is to smooth some noisy measurement data without messing
> up it's *shape*.  My first attempt was 1d splines.  I did:
> 
> from Numeric import *
> import scipy
> f = file('rough')
> data = array([(float(x),float(y)) for (x,y) in [l.split() for l in f]])
> rep = scipy.interpolate.splrep(data[:,0], data[:,1], s=s)
> smooth_y = scipy.interpolate.splev(data[:,0], rep)
> 
> "s" is supposed to vary the amount of smoothing.  For s=0, i get
> my original data back, as expected.  But for all other values, including
> the recommended range of (m-sqrt(2.0*m)) to (m+sqrt(2.0*m), I get a
> single, too much smoothed, result.  It seems like splrep is not using
> the "s" value to adjust the splines, except to sense that it's not zero.
> 
> Splines may not be the right method anyway, since they tend to warp the
> shape of the curve, and I need to get the data's derivatives.  Is there
> some way to fashion a low pass filter?  It seems like fft should be useful
> here, but I have very little experience with fft's.
> 

[snip, snip]

Have you considered wavelet smoothing or denoising?

The 'a-trous' wavelet method is a simple, understandable technique that can be 
done by convolving a function with the data.  The difference between your raw 
data and your smoothed (convolved) data is the wavelet at that particular 
spatial resolution.  You can filter the wavelet values using thresholding and 
then add them back into the smoothed data to get your smoothed or denoised 
result. This can be done at different spatial resolutions depending on the what 
scale you wish to smooth.

If you set the filtering threshold correctly, you'll be able to keep the 
significant features of the data without loosing resolution as is common with 
averaging techniques, or adding spurious features as is common with FFT 
techniques.  A B3-spline is good kernel function and usually does a good job in 
this situation.  Its 1-dimensional representation is [1, 4, 6, 4, 1]/16.

  -- Paul

-- 
Paul Barrett, PhD      Space Telescope Science Institute
Phone: 410-338-4475    ESS/Science Software Branch
FAX:   410-338-4767    Baltimore, MD 21218


From prabhu at aero.iitm.ernet.in  Thu Feb 26 12:06:21 2004
From: prabhu at aero.iitm.ernet.in (Prabhu Ramachandran)
Date: Thu, 26 Feb 2004 22:36:21 +0530
Subject: [SciPy-user] weave, including headers in generated file
In-Reply-To: <Pine.LNX.4.58.0402261450130.24066@nelja.ifi.uio.no>
References: <mailman.0.1076943290.357.scipy-user@scipy.net>
	<Pine.LNX.4.58.0402261450130.24066@nelja.ifi.uio.no>
Message-ID: <16446.10125.224440.819230@monster.linux.in>

>>>>> "MW" == Magne Westlie <magnew at math.uio.no> writes:

    MW> I've got a class Hello declared in a file hw.h, but how do I
    MW> tell weave to include this file in the generated cpp-file? If
    MW> I try:

    MW> code = """ Hello h; h.hw(); """

    MW> weave.inline(code,
    MW>              [], compiler='gcc', verbose=2, force=1,
    MW>              sources=['/home/mw/weave/inline-ex/hw.cpp'],
    MW>              include_dirs=['/home/mw/weave/inline-ex/'], )

Use the headers keyword and put hw.h into it.  Also add the path to
your header in include_dirs.

cheers,
prabhu


From jbenson at nofs.navy.mil  Thu Feb 26 12:12:13 2004
From: jbenson at nofs.navy.mil (James A. Benson)
Date: Thu, 26 Feb 2004 10:12:13 -0700 (MST)
Subject: [SciPy-user] scipy cvs snapshot problem (solved)
In-Reply-To: <Pine.LNX.4.44.0402260233520.31709-100000@scipy.org>
Message-ID: <Pine.LNX.4.44.0402261010110.15133-100000@clavius.nofs.navy.mil>

On Thu, 26 Feb 2004, Pearu Peterson wrote:

> 
> On Wed, 25 Feb 2004, Jim Benson wrote:
> 
> > I downloaded a cvs snapshot on 2004-02-25 and installed it
> > over a previous cvs snapshot (from 2003-12-03). The previous 
> > one did pass all scipy.test levels. The recent version
> > failed the first test:
> > 
> > ======================================================================
> > FAIL: check_nils_20Feb04 (scipy.linalg.basic.test_basic.test_solve)
> > ----------------------------------------------------------------------
> 
> > ...am i forgetting something critical?
> 
> Yes, you missed recent thread about the corner case where linalg.solve 
> could return wrong results. The above test covers this case.
> You should certainly update f2py to the latest if somewhere in your 
> applications the following conditions may be holding:


Thanks! I updated to F2Py-2.39.235_1644, reinstalled the
scipy cvs snapshot, and it now passed all the tests.
                                                                                                     
Thanks,
                                                                                                     
Jim



From gry at ll.mit.edu  Thu Feb 26 13:20:36 2004
From: gry at ll.mit.edu (george young)
Date: Thu, 26 Feb 2004 13:20:36 -0500
Subject: [SciPy-user] 	Re: data smoothing: interpolate.splrep ignores s
	parameter
In-Reply-To: <403DFF9E.20403@stsci.edu>
References: <20040225142126.096ed95e.gry@ll.mit.edu>
	<403DFF9E.20403@stsci.edu>
Message-ID: <20040226132036.436b4c05.gry@ll.mit.edu>

On Thu, 26 Feb 2004 09:15:58 -0500
Paul Barrett <barrett at stsci.edu> threw this fish to the penguins:

> george young wrote:
> > [[reposting with "rough" data file appended, sorry]]
> > [SciPy-0.2.0_alpha_200.4161, Numeric-23.1, Python 2.3.3, x86 Linux]
> > My goal really is to smooth some noisy measurement data without messing
> > up it's *shape*.  My first attempt was 1d splines.  I did:
> > 
...
> > Splines may not be the right method anyway, since they tend to warp the
> > shape of the curve, and I need to get the data's derivatives.  Is there
> > some way to fashion a low pass filter?  It seems like fft should be useful
> > here, but I have very little experience with fft's.
> > 
> 
> [snip, snip]
> 
> Have you considered wavelet smoothing or denoising?
> 
> The 'a-trous' wavelet method is a simple, understandable technique that can be 
> done by convolving a function with the data.  The difference between your raw 
> data and your smoothed (convolved) data is the wavelet at that particular 
> spatial resolution.  You can filter the wavelet values using thresholding and 
> then add them back into the smoothed data to get your smoothed or denoised 
> result. This can be done at different spatial resolutions depending on the what 
> scale you wish to smooth.
> 
> If you set the filtering threshold correctly, you'll be able to keep the 
> significant features of the data without loosing resolution as is common with 
> averaging techniques, or adding spurious features as is common with FFT 
> techniques.  A B3-spline is good kernel function and usually does a good job in 
> this situation.  Its 1-dimensional representation is [1, 4, 6, 4, 1]/16.

That works very nicely, thanks!  It decreases the noise locally without
warping the overall shape.  I just did: 

    ytmp = zeros(len(data[:, 1]), Float)
    xmin=0; xmax=len(data)-1
    fn= [[-2,1], [-1,4], [0,6], [1,4], [2,1]]
    for i in range(len(data)):
        y = w = 0.0
        for f in fn:
            if xmin <= (i + f[0]) <= xmax:
                y += f[1] * data[i + f[0], 1]
                w += f[1]
        ytmp[i] = y / w

    data[:, 1] = ytmp

I know nothing of wavelets or dsp.  Is there something I could read in a few hours
to understand a little of "a-trous" and "kernels"?  

Is it safe to play with the numbers, e.g. [1,5,8,5,1]/20 or is there 
subtlety that will bite me?

I assume there's functionality in scipy for doing this instead of hard
coding it?  I noticed a convolve function, but not much documentation.


Thanks very much again,
	George Young
 
-- 
"Are the gods not just?"  "Oh no, child.
What would become of us if they were?" (CSL)


From oliphant at ee.byu.edu  Thu Feb 26 17:38:17 2004
From: oliphant at ee.byu.edu (Travis Oliphant)
Date: Thu, 26 Feb 2004 15:38:17 -0700
Subject: [SciPy-user] FIR Filters
In-Reply-To: <9B40DB5FBF7D2048AD79C2A91F862A51772F68@STEVAPPS05.Stevenage.CGI-Europe.com>
References: <9B40DB5FBF7D2048AD79C2A91F862A51772F68@STEVAPPS05.Stevenage.CGI-Europe.com>
Message-ID: <403E7559.6000309@ee.byu.edu>

Bob.Cowdery at CGI-Europe.com wrote:

> Travis
>
> Thanks very much for the responses. Don't confuse me with someone who 
> knows what they are doing (but I am working on it!). To give you some 
> background I am converting a .NET application that uses Intel IPP to 
> Python. The application is for a software defined radio. At the moment 
> I have audio capture which contains my quadrature samples in blocks of 
> 4096 (2048 each for I&Q) with 44100 sampling rate (46ms per block). I 
> have a simple dsp process to test each signal processing call. At the 
> moment this should output exactly what it receives, i.e just the audio 
> input as the demodulator needs the FIR filter and some magic applied 
> at the point indicated in the code. When I have written the 
> demodulator I can test firwin() properly. At the moment I can see it 
> outputs valid data, and it's fast which is good. The code below is the 
> very simple path. I have had to write two functions for 
> polar/cartesian conversions as I didn't find any in the library. I 
> don't know if these work right as I can hear clicking sounds on the 
> audio not there when they are commented out. However, it could be 
> audio overrun as I am up at 75% processor. The functions are also 
> quite slow (adding 30% to
>
> my processor loading). Is there a more efficient way to do this?
>

All the best in your education...

First,

Are you using numarray?  if so then that might cause a slower 
application than using Numeric (older but often faster arraymodule)

Comments on your code below:

>     def sigproc( self, samples ) :       
>        
>         # convert to a numarray array of type double
>         x = asarray(samples, typecode=Float64)
>
            is this numarray.asarray or Numeric.asarray ?

>        
>         # first to do is sort into real and image arrays from 
> interleaved array
>         realin = x[1::2]  # every other element starting at index 1
>         imagin = x[::2]   # every other element starting at index 0 \
>
>                
>         # convert to frequency domain
>         realfd = fft(realin)
>         imagfd = fft(imagin)
>

          Normally, real and imaginary parts of the data are handled 
together when a complex data type is available.    I would do
          in = x[1::2] + 1j*x[::2]
          fd = fft(in)

>        
>         # cartesian to polar in frequency domain
>         magn,phase = self.cartesian_to_polar(realfd, imagfd)
>        
>
          This seems strange to me, why are you filtering the magnitude 
and phase separately?  This is very non-standard.

>
>         # apply filter and demodulate
>
           I doubt you want a 2048 tap filter at this point.  Probably 
something a lot less.
           What kind of modulation are you demodulating?   Are you doing 
frequency-domain filtering then?

>                
>         # polar to cartesian in frequency domain
>
>         realfd, imagfd = self.polar_to_cartesian(magn, phase)
>

>        
>         # convert to time domain
>         realtd = ifft(realfd)
>         imagtd = ifft(imagfd)
>                
>         # convert back to interleaved array
>         x[::2] = realtd.astype(Float64)
>         x[1::2] = imagtd.astype(Float64)
>        
>         # convert back to an integer array
>         x = x.astype('l')
>        
>         # return python list of processed samples
>         return x.tolist()
>
     Why are you going back to a python list?  A numeric array is a 
valid sequence.  Why does it need to be a list.  You have a lot of 
memory copying going on which is going to slow down everything.


>    
>     # utility functions not in the library
>     def polar_to_cartesian(self, magn, phase):
>         return magn * cos(phase), magn * sin(phase)
>
>     def cartesian_to_polar(self, real, imag):
>         # convert the arrays real, imag into the polar system
>         # magnitude array
>         magn = sqrt(real**2 + imag**2)
>         # phase array
>         phase = 
> choose(equal(real,0),(arctan(imag/real),(choose(less(imag,0),((pi/2),-(pi/2)))))) 
>
>         return magn, phase
>
>

If you have a complex-valued array, then abs(x) is the magnitude and 
angle(x) is the phase.

My colleagues down the hall do software radio and I have a lot of 
background in signal processing.  Let me know if you have more specific 
questions.

Thanks,

-Travis O.



From paxcalpt at sapo.pt  Thu Feb 26 18:31:15 2004
From: paxcalpt at sapo.pt (Ricardo Henriques)
Date: Thu, 26 Feb 2004 23:31:15 -0000
Subject: [SciPy-user] Boa Constructor Plug-in for ploting
Message-ID: <003501c3fcc0$a28fa870$5181c151@xcal>

Is there a Boa Constructor plug-in using a scipy based plot interface like
Chaco or scipy.plt out there?
I have been using wxPyPlot witch is based on scipy.plot but laks many of its
features...

Tks...
Paxcal


From pajer at iname.com  Thu Feb 26 23:11:30 2004
From: pajer at iname.com (Gary Pajer)
Date: Thu, 26 Feb 2004 23:11:30 -0500
Subject: [SciPy-user] Boa Constructor Plug-in for ploting
References: <003501c3fcc0$a28fa870$5181c151@xcal>
Message-ID: <001901c3fce7$c741d090$01fd5644@playroom>

----- Original Message ----- 
From: "Ricardo Henriques" <paxcalpt at sapo.pt>
> Is there a Boa Constructor plug-in using a scipy based plot interface like
> Chaco or scipy.plt out there?
> I have been using wxPyPlot witch is based on scipy.plot but laks many of
its
> features...

I'll second the question, and add something: I've been experimenting with
BLT/Tkinter, and I'm growing fond of it.  Lots of features, including a
stripchart widget that works for me nicely.

-gary


From arnd.baecker at web.de  Fri Feb 27 05:12:09 2004
From: arnd.baecker at web.de (Arnd Baecker)
Date: Fri, 27 Feb 2004 11:12:09 +0100 (CET)
Subject: [SciPy-user] help and lazy-import (?)
Message-ID: <Pine.LNX.4.51.0402271111070.32377@ptpcp2.phy.tu-dresden.de>

Hi,

there there seems to be something weird
going on when trying to access help in the following way

>>> import scipy
>>> help("scipy.linalg")
Help on instance of _ModuleLoader in scipy:

scipy.linalg = <module 'scipy.linalg' from '/opt/python/lib/pyt...
e-packages/scipy/linalg/__init__.pyc' [imported]>

Doing another
>>> help("scipy.linalg")

one gets the expected help.

I'd guess that this somehow has to to with the lazy importer of scipy.
But maybe there is a workaround?

Many thanks in advance,

Arnd


From pearu at scipy.org  Fri Feb 27 06:08:01 2004
From: pearu at scipy.org (Pearu Peterson)
Date: Fri, 27 Feb 2004 05:08:01 -0600 (CST)
Subject: [SciPy-user] help and lazy-import (?)
In-Reply-To: <Pine.LNX.4.51.0402271111070.32377@ptpcp2.phy.tu-dresden.de>
Message-ID: <Pine.LNX.4.44.0402270450590.31709-100000@scipy.org>



On Fri, 27 Feb 2004, Arnd Baecker wrote:

> there there seems to be something weird
> going on when trying to access help in the following way
> 
> >>> import scipy
> >>> help("scipy.linalg")
> Help on instance of _ModuleLoader in scipy:
> 
> scipy.linalg = <module 'scipy.linalg' from '/opt/python/lib/pyt...
> e-packages/scipy/linalg/__init__.pyc' [imported]>
> 
> Doing another
> >>> help("scipy.linalg")
> 
> one gets the expected help.
> 
> I'd guess that this somehow has to to with the lazy importer of scipy.
> But maybe there is a workaround?

I get similar behaviour when using ipython. But in a traditional
Python prompt

>>> import scipy
>>> help("scipy.linalg")

I get expected help. This is because ppimport redefines built-in help:

>>> help
Type help () for interactive help, or help (object) for help about object.
>>> help.__module__
'scipy_base.ppimport'

So, for some reasons, in my case when using ipython, redefined help is 
not effective and probably because help is referenced in ipython before 
importing scipy. It seems that you are not using ipython, so, what is the 
contents of your PYTHONSTARTUP?

Pearu


From arnd.baecker at web.de  Fri Feb 27 06:24:51 2004
From: arnd.baecker at web.de (Arnd Baecker)
Date: Fri, 27 Feb 2004 12:24:51 +0100 (CET)
Subject: [SciPy-user] help and lazy-import (?)
In-Reply-To: <Pine.LNX.4.44.0402270450590.31709-100000@scipy.org>
References: <Pine.LNX.4.44.0402270450590.31709-100000@scipy.org>
Message-ID: <Pine.LNX.4.51.0402271206240.32377@ptpcp2.phy.tu-dresden.de>


On Fri, 27 Feb 2004, Pearu Peterson wrote:

> On Fri, 27 Feb 2004, Arnd Baecker wrote:
>
> > there there seems to be something weird
> > going on when trying to access help in the following way
> >
> > >>> import scipy
> > >>> help("scipy.linalg")
> > Help on instance of _ModuleLoader in scipy:
> >
> > scipy.linalg = <module 'scipy.linalg' from '/opt/python/lib/pyt...
> > e-packages/scipy/linalg/__init__.pyc' [imported]>
> >
> > Doing another
> > >>> help("scipy.linalg")
> >
> > one gets the expected help.
> >
> > I'd guess that this somehow has to to with the lazy importer of scipy.
> > But maybe there is a workaround?
>
> I get similar behaviour when using ipython.

Well, that's where I first saw this as well
(honestly, I always use ipython ;-)
As I also see this with the normal python prompt
I thought it is un-related to ipython.

> But in a traditional Python prompt
>
> >>> import scipy
> >>> help("scipy.linalg")
>
> I get expected help. This is because ppimport redefines built-in help:
>
> >>> help
> Type help () for interactive help, or help (object) for help about object.
> >>> help.__module__
> 'scipy_base.ppimport'

I get precisely the same output
((to be clear here: apart from that help("scipy.linalg") gives
the above _ModuleLoader message on the first invocation, and only the
second time I get the expected help.))

If this matters:
  - scipy.__version__: '0.2.1_253.3701'
  - Python: 2.3.3

> So, for some reasons, in my case when using ipython, redefined help is
> not effective and probably because help is referenced in ipython before
> importing scipy.

With IPython 0.5.1.cvs I get the same as described above, however with
In [4]: help.__module__
Out[4]: 'site'

> It seems that you are not using ipython, so, what is the
> contents of your PYTHONSTARTUP?

no contents, I don't use it.

What further tests could I do solve this issue ?
(maybe we need further tests by other users to find the origin?)

Many thanks,

Arnd


P.S.: just another "screen shot":

Python 2.3.3 (#1, Feb  3 2004, 22:05:11)
[GCC 3.2] on linux2
Type "help", "copyright", "credits" or "license" for more information.
>>> help.__module__
'site'
>>> import scipy
>>> help.__module__
'scipy_base.ppimport'
>>> help("scipy.linalg")
Help on instance of _ModuleLoader in scipy:

scipy.linalg = <module 'scipy.linalg' from
'/opt/python/lib/pyt...e-packages/sci
py/linalg/__init__.pyc' [imported]>
>>> help("scipy.linalg")

Help on package scipy.linalg in scipy:

NAME
    scipy.linalg

FILE
[..]


From Bob.Cowdery at CGI-Europe.com  Fri Feb 27 09:10:39 2004
From: Bob.Cowdery at CGI-Europe.com (Bob.Cowdery at CGI-Europe.com)
Date: Fri, 27 Feb 2004 14:10:39 -0000
Subject: [SciPy-user] FIR Filters
Message-ID: <9B40DB5FBF7D2048AD79C2A91F862A51772F6B@STEVAPPS05.Stevenage.CGI-Europe.com>

Travis

With your help I am making progress. I started off using numarray, then ran
into problems so I am now using Numeric. I only import scipy so I am using
whatever it imports which I believe is Numeric. I have tried to follow your
advice and move to complex types as I can see it makes sense. I don't think
I have done a complete job yet. The audio output has a pulse type modulation
on it which should not be there, however selecting different passbands is
definitely having the desired effect. I have copied all the code in below
for context. Yes the filtering is in the frequency domain which I am told
gives better stopband attenuation and good shape. The way the filter and
demodulation works is still a bit of a mystery to me as I didn't write the
original code. I would love to work out exactly what it does but I'm not
quite there yet. I can't really ask very targetted questions yet until I
understand more. The code cleary still has problems but I don't know why or
where. The good news is the processor is down to 50% now from 80%. If you do
have a few moments perhaps you could point out any more obvious errors.

Thanks
Bob

import time
import threading
import fastaudio as f
from scipy import *

class dsp( threading.Thread ):

    # signal processing
    def sig( self, samples ) :
        
        # allocate the filter arrays
        global m_filterMagn
        global m_filterPhase
        
        # convert to a numeric array of type double
        x = asarray(samples, typecode=Float64)
        
        # get a complex array ordered from the I,Q source
        complexin = x[1::2] + 1j*x[::2]
        
        # convert to complex frequency domain
        cart_complex_in_fd = fft(complexin)
        
        # apply filter to magnitude and phase
        magn = abs(complexin)*m_filterMagn
        phase = angle(complexin)*m_filterPhase
        
        # polar to cartesian in frequency domain
        demodulated = magn * cos(phase) + 1j*(magn * sin(phase))
        
        # convert to time domain
        complexout = ifft(demodulated)
        
        # convert back to interleaved array
        x[1::2] = real(complexout)
        x[::2] = real(complexout)
        
        # convert back to an integer array
        x = x.astype('l')
        
        # return sequence of processed samples
        return x
    
    def firwin(self, N, cutoff, width=None, window='blackman'):
        """
        FIR Filter Design using windowed ideal filter method.

        Inputs:
   
          N      -- order of filter (number of taps)
          cutoff -- cutoff frequency of filter 
                      (normalized so that 1 corresponds to Nyquist or pi
radians / sample)
          width  -- if width is not None, then assume it is the approximate
width of
                      the transition region (normalized so that 1 corresonds
to pi)
                      for use in kaiser FIR filter design.
          window -- desired window to use.

        Outputs:

          h      -- coefficients of length N fir filter.
        """

        if isinstance(width,float):
            A = 2.285*N*width + 8
            if (A < 21): beta = 0.0
            elif (A <= 50): beta = 0.5842*(A-21)**0.4 + 0.07886*(A-21)
            else: beta = 0.1102*(A-8.7)
            window=('kaiser',beta)

        win = signal.get_window(window,N,fftbins=1)
        alpha = N//2
        m = signal.Num.arange(0,N)
        return win*special.sinc(cutoff*(m-alpha))
    
    # calculate the complex filter
    def filter_calc( self, fLow, fHigh ):
    
        global m_filterMagn
        global m_filterPhase

        m_filterMagn = zeros(2048)
        m_filterPhase = zeros(2048)
        
        # create empty arrays to hold results
        Rh = zeros(2048)
        Ih = zeros(2048)
        
        # calculate the high/low cutoff freq as a fraction of the sample
rate
        fhc = fHigh/44100.0
        flc = fLow/44100.0
        
        # calculate fft bin size
        bin_size = 44100.0/4096.0
        
        # calculate no of bins in filter width
        bins = fHigh/bin_size
        
        # calculate FIR lowpass filter coefficients
        Rh = self.firwin(2048.0,(fhc-flc)/2.0)
        
        # compute USB filter by modulating the lowpass filter up
        filterPhase = 0.0
        filterFreq = (flc+fhc)*pi
        phaseSin = 0
        phaseCos = 0

        # modulate to USB centre frequency
        i = 0
        while i< 2048 :
            phaseSin = sin(filterPhase)
            phaseCos = cos(filterPhase)
            Ih[i] = Rh[i]*phaseSin
            Rh[i] = Rh[i]*phaseCos
            filterPhase = filterPhase+filterFreq
            if filterPhase >= 6.28318530717959 :
                filterPhase = filterPhase - 6.28318530717959
            i = i+1
            
        # for LSB create conjugate of USB filter
        # comment out for LSB
        Ih = -Ih
        
        # compute complex frequency domain response of LSB or USB filter
        complexin = Rh + 1j*Ih
        complexfd = fft(complexin)
        
        # Cartesian to polar gives the magnitude and phase components of the
fft result
        m_filterMagn = abs(complexfd)
        m_filterPhase = angle(complexfd)
                  
        
    # start streaming
    def run( self ):

        global running
        running = 1
        
        #s = f.stream(7500,       # sample rate
        #             2,          # number of channels
        #             'int16',    # sample format, choose from 'int8',
'int16', 'int32'.
        #             4096,       # frames per buffer
        #             16)         # maximum number of input buffers

        # open stream for input/output
        print "Create, open and start stream"
        global s
        s = f.stream(44100,2,'int16',2048,4)

        # open and start streaming
        s.open()
        s.start()

        print "Thread running..."
        
        # calculate the SSB filter
        self.filter_calc(300.0,3000.0)
        
        while( running ):

            # read next block of data from the input buffer
            samples = s.read()

            if( len( samples ) > 0 ):
                demod = self.sig( samples )
                # now, write the data back
                s.write( demod )

            # sleep for 10 ms
            # a stero block of 4096 samples takes 46 ms so 10ms should avoid
dropped blocks
            time.sleep(0.01)
            
        print "Thread exiting..."

def start():
    global DSP
    DSP = dsp()
    DSP.start()
    
def stop():
    # stop the stream
    global running
    running = 0
    global s
    s.stop()
    # and close it
    s.close()
    
def filter(id):
    # set the filter passband
    global DSP
    if id == 0:
        DSP.filter_calc(300.0,3000.0)
    elif id == 1:
        DSP.filter_calc(300.0,2000.0)
    elif id == 2:
        DSP.filter_calc(500.0,700.0)  

-----Original Message-----
From: Travis Oliphant [mailto:oliphant at ee.byu.edu] 
Sent: 26 February 2004 22:38
To: SciPy Users List
Subject: Re: [SciPy-user] FIR Filters


Bob.Cowdery at CGI-Europe.com wrote:

> Travis
>
> Thanks very much for the responses. Don't confuse me with someone who
> knows what they are doing (but I am working on it!). To give you some 
> background I am converting a .NET application that uses Intel IPP to 
> Python. The application is for a software defined radio. At the moment 
> I have audio capture which contains my quadrature samples in blocks of 
> 4096 (2048 each for I&Q) with 44100 sampling rate (46ms per block). I 
> have a simple dsp process to test each signal processing call. At the 
> moment this should output exactly what it receives, i.e just the audio 
> input as the demodulator needs the FIR filter and some magic applied 
> at the point indicated in the code. When I have written the 
> demodulator I can test firwin() properly. At the moment I can see it 
> outputs valid data, and it's fast which is good. The code below is the 
> very simple path. I have had to write two functions for 
> polar/cartesian conversions as I didn't find any in the library. I 
> don't know if these work right as I can hear clicking sounds on the 
> audio not there when they are commented out. However, it could be 
> audio overrun as I am up at 75% processor. The functions are also 
> quite slow (adding 30% to
>
> my processor loading). Is there a more efficient way to do this?
>

All the best in your education...

First,

Are you using numarray?  if so then that might cause a slower 
application than using Numeric (older but often faster arraymodule)

Comments on your code below:

>     def sigproc( self, samples ) :       
>        
>         # convert to a numarray array of type double
>         x = asarray(samples, typecode=Float64)
>
            is this numarray.asarray or Numeric.asarray ?

>        
>         # first to do is sort into real and image arrays from
> interleaved array
>         realin = x[1::2]  # every other element starting at index 1
>         imagin = x[::2]   # every other element starting at index 0 \
>
>                
>         # convert to frequency domain
>         realfd = fft(realin)
>         imagfd = fft(imagin)
>

          Normally, real and imaginary parts of the data are handled 
together when a complex data type is available.    I would do
          in = x[1::2] + 1j*x[::2]
          fd = fft(in)

>        
>         # cartesian to polar in frequency domain
>         magn,phase = self.cartesian_to_polar(realfd, imagfd)
>        
>
          This seems strange to me, why are you filtering the magnitude 
and phase separately?  This is very non-standard.

>
>         # apply filter and demodulate
>
           I doubt you want a 2048 tap filter at this point.  Probably 
something a lot less.
           What kind of modulation are you demodulating?   Are you doing 
frequency-domain filtering then?

>                
>         # polar to cartesian in frequency domain
>
>         realfd, imagfd = self.polar_to_cartesian(magn, phase)
>

>        
>         # convert to time domain
>         realtd = ifft(realfd)
>         imagtd = ifft(imagfd)
>                
>         # convert back to interleaved array
>         x[::2] = realtd.astype(Float64)
>         x[1::2] = imagtd.astype(Float64)
>        
>         # convert back to an integer array
>         x = x.astype('l')
>        
>         # return python list of processed samples
>         return x.tolist()
>
     Why are you going back to a python list?  A numeric array is a 
valid sequence.  Why does it need to be a list.  You have a lot of 
memory copying going on which is going to slow down everything.


>    
>     # utility functions not in the library
>     def polar_to_cartesian(self, magn, phase):
>         return magn * cos(phase), magn * sin(phase)
>
>     def cartesian_to_polar(self, real, imag):
>         # convert the arrays real, imag into the polar system
>         # magnitude array
>         magn = sqrt(real**2 + imag**2)
>         # phase array
>         phase =
>
choose(equal(real,0),(arctan(imag/real),(choose(less(imag,0),((pi/2),-(pi/2)
))))) 
>
>         return magn, phase
>
>

If you have a complex-valued array, then abs(x) is the magnitude and 
angle(x) is the phase.

My colleagues down the hall do software radio and I have a lot of 
background in signal processing.  Let me know if you have more specific 
questions.

Thanks,

-Travis O.


_______________________________________________
SciPy-user mailing list
SciPy-user at scipy.net http://www.scipy.net/mailman/listinfo/scipy-user

*** Confidentiality Notice *** Proprietary/Confidential
Information belonging to CGI Group Inc. and its affiliates
may be contained in this message. If you are not a recipient
indicated or intended in this message (or responsible for
delivery of this message to such person), or you think for
any reason that this message may have been addressed to you
in error, you may not use or copy or deliver this message
to anyone else.  In such case, you should destroy this
message and are asked to notify the sender by reply email.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.scipy.org/pipermail/scipy-user/attachments/20040227/d67b11e1/attachment.html>

From Fernando.Perez at colorado.edu  Fri Feb 27 12:21:31 2004
From: Fernando.Perez at colorado.edu (Fernando Perez)
Date: Fri, 27 Feb 2004 10:21:31 -0700
Subject: [SciPy-user] help and lazy-import (?)
In-Reply-To: <Pine.LNX.4.44.0402270450590.31709-100000@scipy.org>
References: <Pine.LNX.4.44.0402270450590.31709-100000@scipy.org>
Message-ID: <403F7C9B.4030909@colorado.edu>

Pearu Peterson wrote:

> So, for some reasons, in my case when using ipython, redefined help is 
> not effective and probably because help is referenced in ipython before 
> importing scipy. It seems that you are not using ipython, so, what is the 
> contents of your PYTHONSTARTUP?

Indeed, ipython imports help upon starting.  So by the time scipy tries to 
modify things, the help ipython sees is already the default one.

I've noticed in general that, because of the lazy import mechanisms of scipy, 
often I need to do things like hitting tab twice, at least the first time I 
try them.  I realize the rationale for the lazy import system, so I don't know 
if this is something which can even be fixed at all.  But if there are any 
changes which ipython can make to improve the situation, and someone can at 
least point me in the right direction, I'll gladly give it a try.

Regards,

f.