From scipy at zunzun.com  Sun May  2 17:44:00 2004
From: scipy at zunzun.com (James R. Phillips)
Date: Sun, 2 May 2004 17:44:00 -400
Subject: [SciPy-user] Unable to use weave in Enthought 0.3 Win binary install
Message-ID: <40956ba02e07f7.15773117@mercury.sabren.com>

Has anyone had success using weave in the
Enthought 0.3 Winows binary install?  I get
linker errors whether using the Enthought-supplied
Mingw32 or my MSVC++ 6.0 compiler.

My test

I have been through several install-reboot-test
cycles on two different XP boxen.  Arrgh.

If I knew the magic chants to get weave working
with Enthought's Mingw32 it would be sufficient.

     James Phillips
     http://zunzun.com



From danny_shevitz at yahoo.com  Mon May  3 19:07:41 2004
From: danny_shevitz at yahoo.com (danny shevitz)
Date: Mon, 3 May 2004 16:07:41 -0700 (PDT)
Subject: [SciPy-user] problems with stats
In-Reply-To: <408EDC12.9080005@ucsd.edu>
Message-ID: <20040503230741.45620.qmail@web41013.mail.yahoo.com>

Robert, 

thanks for the response. I haven't forgotten you wrote, I just haven't
had a chance to really dig into what you wrote. Now I have. 

So I have been using the frequentist approach to constructing the error
bars and I understand that. It involves taking some linear fractional
transformations of F statistics. After reading your letter, I was able
to compute the same error bounds directly. The key was the bdtri
function as you mentioned. 

pLower = bdtri(n-1, N, .95)
pUpper = bdtri(n, N, .05)

for example gives the .90 confidence interval. This is the pair of
functions I needed. The n-1 in pLower is because the value n is
considered in the tail for the purposes of computing the confidences.

I didn't have a prayer in scipy.stats because bdtri is not wrapped. Its
totally different than ppf, which still requires the unknown
probability in order to construct the distribution at all. 

So I can get the frequentist result directly, now I'm curious about
your Bayesian analysis. One of the uniform or improper prior gives the
same answer. BTW, I'm also curious if you left off the combinatoric
factor (n choose r) in your description. Can you give me a little of
the rational for choosing the improper prior?

thanks,
Danny


--- Robert Kern <rkern at ucsd.edu> wrote:
> danny shevitz wrote:
> 
> > The short version of the problem that I was trying to solve is:
> Given N
> > trials and n events, what is the specified confidence interval for
> the
> > binomial probability p.
> > 
> > The hypothetical argument call should therefore take 2 integers and
> a
> > probability, e.g. f(10, 10000000, .95) The third number is a
> > confidence, not the probability in the binomial distribution. Of
> course
> > there would have to be a pair of such functions ( one for the lower
> > bound, and one for the upper). 
> 
> What you are looking for is the beta distribution. Here's the
> Bayesian 
> analysis:
> 
> We have n Bernoulli trials and r successes with an unknown chance 
> (theta) of success on any given trial. The likelihood is
> 
> P(n,r|theta) = theta**r * (1-theta)**(n-r)
> 
> The posterior probability is
> 
> P(theta|n,r) = A * P(n,r|theta) * P(theta)
> 
> A is a normalizing constant and P(theta) is your prior probability
> for 
> theta. For now, let's use a minimally informative prior in order to
> make 
> the fewest assumptions.
> 
> P(theta) o< 1/(theta*(1-theta))
> 
> The o< symbol is the proportionality symbol. This prior is improper
> and 
> cannot be normalized as it stands. When we put it together with the 
> likelihood, as long as r does not equal either 0 or n, the posterior 
> will come out normalized. There are good reasons which I won't go
> into 
> now to use this instead of P(theta)=1, but you can use that instead
> and 
> redo the math that leads to the following (easy!).
> 
> So, P(theta|n,r) = A * theta**(r-1) * (1-theta)**(n-r-1)
> 
> We can recognize this as the Beta distribution, Beta(r,n-r) (or 
> Beta(r+1,n-r+1) if you're queasy about improper priors).
> 
> special.bdtri(10, 10000000, 0.95) will give you the one-sided 95% 
> credible interval (like a confidence interval, only Bayesian). Since 
> this distribution isn't symmetric, I'm not entirely sure how to get
> the 
> highest posterior density (HPD) credible interval around the peak.
> And 
> I'm not entirely sure how accurate the function is with those extreme
> 
> inputs.
> 
> This analysis is lifted straight from Chapter 6 of the excellent book
> 
> _Probability Theory: The Logic of Science_ by E. T. Jaynes (Cambridge
> 
> Press, 2003).
> 
> > Like I said, what I was doing, wasn't particularly well thought
> out. I
> > was just playing around, and using the function wrong anyway...
> > 
> > BTW I guess I have no idea what binom.ppf is supposed to do then.
> Why
> > do you need two probilities? Is one a confidence and the other the
> > probability.
> 
> The random variable in the binomial distribution is the number of 
> successes given the probability p of individual success and the
> number 
> of trials. The point mass function would invert for n, the number of 
> successes that, by summing the probability masses for each number of 
> successes between 0 and n, gives you 0.95 (for example). So yes, one 
> input is the confidence level, one is the probability of individual 
> success, and you also need the number of trials.
> 
> > Also in case it wasn't obvious, I'm not a statistician by training
> :-)
> > 
> > D  
> 
> -- 
> 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
> 
> _______________________________________________
> SciPy-user mailing list
> SciPy-user at scipy.net
> http://www.scipy.net/mailman/listinfo/scipy-user



	
		
__________________________________
Do you Yahoo!?
Win a $20,000 Career Makeover at Yahoo! HotJobs  
http://hotjobs.sweepstakes.yahoo.com/careermakeover 



From danny_shevitz at yahoo.com  Mon May  3 19:10:54 2004
From: danny_shevitz at yahoo.com (danny shevitz)
Date: Mon, 3 May 2004 16:10:54 -0700 (PDT)
Subject: [SciPy-user] problems with stats
In-Reply-To: <20040503230741.45620.qmail@web41013.mail.yahoo.com>
Message-ID: <20040503231054.92598.qmail@web41012.mail.yahoo.com>

sorry, for the post, I meant to send it to Robert not the group.

Danny




	
		
__________________________________
Do you Yahoo!?
Win a $20,000 Career Makeover at Yahoo! HotJobs  
http://hotjobs.sweepstakes.yahoo.com/careermakeover 



From nwagner at mecha.uni-stuttgart.de  Tue May  4 02:26:33 2004
From: nwagner at mecha.uni-stuttgart.de (Nils Wagner)
Date: Tue, 04 May 2004 08:26:33 +0200
Subject: [SciPy-user] scipy.test() failed with latest cvs
Message-ID: <40973799.8050207@mecha.uni-stuttgart.de>

======================================================================
ERROR: check_setelement (scipy.sparse.Sparse.test_Sparse.test_csc)
----------------------------------------------------------------------
Traceback (most recent call last):
   File 
"/usr/lib/python2.3/site-packages/scipy/sparse/tests/test_Sparse.py", 
line 67, in check_setelement
     a = self.datsp - self.datsp
   File "/usr/lib/python2.3/site-packages/scipy/sparse/Sparse.py", line 
379, in __sub__
     return csc_matrix(c,(rowc,ptrc),M=M,N=N)
   File "/usr/lib/python2.3/site-packages/scipy/sparse/Sparse.py", line 
294, in __init__
     self._check()
   File "/usr/lib/python2.3/site-packages/scipy/sparse/Sparse.py", line 
310, in _check
     if (nzmax>0) and (max(self.rowind[:nnz]) >= M):
ValueError: min() or max() arg is an empty sequence

----------------------------------------------------------------------
Ran 934 tests in 2.879s

FAILED (errors=1)



From karthik at james.hut.fi  Tue May  4 04:10:53 2004
From: karthik at james.hut.fi (Karthikesh Raju)
Date: Tue, 4 May 2004 11:10:53 +0300
Subject: [SciPy-user] concatenation of arrays woes !!
Message-ID: <Pine.SGI.4.58.0405041055040.5536421@james.hut.fi>


Hi All,

is it possible to concatenate arrays along the column,

A[:,1:5] = concatenate(a,b,c)
a is 2x1, b 2x1, c 1x1

i have been trying to convert the following matlab code and this is my
stumbling block for several months :(

K = 3
CL = 5
L = 2
d -- (K,L) in the range (0,CL)
codes -- (K,CL)
window = 1
s = codes'

output S -(CL*window,K*L*(window+1)) matrix

for k = 1:K
for l = 1:L
    S(:,1+(window+1)*(l-1)+(window+1)*L*(k-1)) =
           [s(CL-d(k,l)+1:CL,k); zeros(CL*(window-1),1);zeros(CL-d(k,l),1)]
    S(:,(window+1)*1+(window+1)*L*(k-1)) =
           [zeros(d(k,l),1);zeros(CL*(window-1),1);s(1:CL-d(k,l),k)]
    for n=2:window
        S(:,n+(window+1)*(l-1)+(numbits+1)*L*(k-1)) =
           [zeros(d(k,l),1); zeros(CL*(n-2),1);
s(:,k);zeros((window-n)*CL+CL-d(k,l),1)]
    end
end
end

--------------------

This module has been my stumbling block for the last 2 months, had no
problems with matlab, here i have all kinds of concatination woes. Please
give me pointers on how to achieve such concatenations,

thanks a lot

regards
karthik





-----------------------------------------------------------------------
Karthikesh Raju,		    email: karthik at james.hut.fi
Researcher,			    http://www.cis.hut.fi/karthik
Helsinki University of Technology,  Tel: +358-9-451 5389
Laboratory of Comp. & Info. Sc.,    Fax: +358-9-451 3277
Department of Computer Sc.,
P.O Box 5400, FIN 02015 HUT,
Espoo, FINLAND
-----------------------------------------------------------------------



From nadavh at visionsense.com  Tue May  4 06:05:30 2004
From: nadavh at visionsense.com (Nadav Horesh)
Date: Tue, 4 May 2004 12:05:30 +0200
Subject: [SciPy-user] Re: xplt documentation
Message-ID: <07C6A61102C94148B8104D42DE95F7E86DECEA@exchange2k.envision.co.il>

I've made some testing (not exhaustive) of matplotlib under IDLE and pycrust on win32 and linux platforms. matplotlib seems to be more tolerant then xplt, especially when show() is called instead of using the interactive mode. The linux environment seems to work a bit better then the win32.

There is a project (not very active nowadays) called glplot (glplot.sf.net) which fills a shortcoming of gnuplot of generating false colour map of large matrices (gnuplot is too slow here). Could the solution be a library/utility that is spawned as a separate process and use some (simple) interprocess communication (ala gnuplot/glplot)? Is there a portable alternative way to run a multiple message-loops under the same process?

  Nadav.



From Bob.Cowdery at CGI-Europe.com  Tue May  4 05:20:37 2004
From: Bob.Cowdery at CGI-Europe.com (Bob.Cowdery at CGI-Europe.com)
Date: Tue, 4 May 2004 10:20:37 +0100 
Subject: [SciPy-user] Help with array functions
Message-ID: <9B40DB5FBF7D2048AD79C2A91F862A51772FFE@STEVAPPS05.Stevenage.CGI-Europe.com>

Hi all,
 
Can anyone help me convert this loop to an array function. I can't have
loops in my signal processing as they take far too much processing power.
This is a fragment from a speech processor using Ferkins formula.
Essentially the gain array (m_procGain) which is 2048 long (m_sampl_sz)
needs its nth element processing into it's (n-1)th element. This is
eventually used to multiply the samples in m_complexout, the first 2048
elements of this are used. Attack is a scaler currently 0.4.
                peak =
abs(self.m_complexout[argmax(abs(self.m_complexout[:self.m_sampl_sz]))])
                i = arange(1,self.m_sampl_sz)
                for x in i:
                    if abs(self.m_complexout[x-1]) > 0:
                       self.m_procGain[x] = ( (self.m_procGain[x-1] * (1 -
attack)) + ((attack * peak) / abs(self.m_complexout[x-1])) )         
                       if self.m_procGain[x] > level:
                           self.m_procGain[x] = level
                    else:
                       self.m_procGain[x] = 1

Thanks for any assistance.
 
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/20040504/33aefcd0/attachment.html>

From arnd.baecker at web.de  Tue May  4 08:49:33 2004
From: arnd.baecker at web.de (Arnd Baecker)
Date: Tue, 4 May 2004 14:49:33 +0200 (CEST)
Subject: [SciPy-user] Re: xplt documentation
In-Reply-To: <07C6A61102C94148B8104D42DE95F7E86DECEA@exchange2k.envision.co.il>
References: <07C6A61102C94148B8104D42DE95F7E86DECEA@exchange2k.envision.co.il>
Message-ID: <Pine.LNX.4.51.0405041440130.12522@ptpcp2.phy.tu-dresden.de>

Hi,


On Tue, 4 May 2004, Nadav Horesh wrote:

> I've made some testing (not exhaustive) of matplotlib under IDLE and
> pycrust on win32 and linux platforms. matplotlib seems to be more
> tolerant then xplt,

Can you be a bit more explicit, i.e. give examples ?


> especially when show() is called instead of
> using the interactive mode.

Personally I think that this is one of the big strengths
of scipy.xplt. For example, when debugging
I use a lot
  from IPython.Shell import IPythonShellEmbed
  ipshell = IPythonShellEmbed()

  ipshell()

The ipshell() can come at any point of your code
and all variables are available.
So a quick
  plg(some_array)
or
  pli(some_2d_array)
often allows to find bugs much more quickly
than anything else.

> The linux environment seems to work a
> bit better then the win32.
>
> There is a project (not very active nowadays) called glplot
> (glplot.sf.net) which fills a shortcoming of gnuplot of
> generating false colour map of large matrices (gnuplot is too slow here).

There is a patch on sourceforge for gnuplot which allows
for bitmap images (ie. matrices). I once tried it out and it is very nice.
It might take a little time until this gets integrated
as the gnuplot team seems to be a bit busy with
after release issues.

If you really want to go for large matrices have a look
at MayaVi,
  http://mayavi.sourceforge.net/
In particular
  http://mayavi.sourceforge.net/docs/guide/x967.html#TOOLS

Best,

Arnd



From pajer at iname.com  Tue May  4 09:26:25 2004
From: pajer at iname.com (Gary Pajer)
Date: Tue, 4 May 2004 09:26:25 -0400
Subject: [SciPy-user] concatenation of arrays woes !!
References: <Pine.SGI.4.58.0405041055040.5536421@james.hut.fi>
Message-ID: <005b01c431db$6a244fe0$01fd5644@playroom>

----- Original Message ----- 
From: "Karthikesh Raju" <karthik at james.hut.fi>
To: <scipy-user at scipy.net>
Sent: Tuesday, May 04, 2004 4:10 AM
Subject: [SciPy-user] concatenation of arrays woes !!


>
> Hi All,
>
> is it possible to concatenate arrays along the column,
>
> A[:,1:5] = concatenate(a,b,c)
> a is 2x1, b 2x1, c 1x1
>
> i have been trying to convert the following matlab code and this is my
> stumbling block for several months :(
>
> K = 3
> CL = 5
> L = 2
> d -- (K,L) in the range (0,CL)
> codes -- (K,CL)
> window = 1
> s = codes'
>
> output S -(CL*window,K*L*(window+1)) matrix
>
> for k = 1:K
> for l = 1:L
>     S(:,1+(window+1)*(l-1)+(window+1)*L*(k-1)) =
>            [s(CL-d(k,l)+1:CL,k);
zeros(CL*(window-1),1);zeros(CL-d(k,l),1)]
>     S(:,(window+1)*1+(window+1)*L*(k-1)) =
>            [zeros(d(k,l),1);zeros(CL*(window-1),1);s(1:CL-d(k,l),k)]
>     for n=2:window
>         S(:,n+(window+1)*(l-1)+(numbits+1)*L*(k-1)) =
>            [zeros(d(k,l),1); zeros(CL*(n-2),1);
> s(:,k);zeros((window-n)*CL+CL-d(k,l),1)]
>     end
> end
> end
>
> --------------------
>
> This module has been my stumbling block for the last 2 months, had no
> problems with matlab, here i have all kinds of concatination woes. Please
> give me pointers on how to achieve such concatenations,
>
> thanks a lot
>
> regards
> karthik



I'm still not sure what you are trying to achieve.  If you post a minimal
contrived example of what you are trying to do instead of a clip from the
middle of your code it would be easier to figure out.

But have you explored r_[ ]  and c_[ ] ?   ('c' for column, 'r' for row)
These constructions are designed to aid concatenation:

>>> from scipy import *
>>> x=array([1,2])
>>> y=array([3,4])

>>> c_[x,y]
array([1, 2, 3, 4])

>>> c_[x[:,NewAxis],y[:,NewAxis]]
array([[1, 3],
       [2, 4]])

>>> r_[x[:,NewAxis],y[:,NewAxis]]
array([[1],
       [2],
       [3],
       [4]])


hth,
gary



From jdhunter at ace.bsd.uchicago.edu  Tue May  4 09:31:55 2004
From: jdhunter at ace.bsd.uchicago.edu (John Hunter)
Date: Tue, 04 May 2004 08:31:55 -0500
Subject: [SciPy-user] Re: xplt documentation
In-Reply-To: <Pine.LNX.4.51.0405041440130.12522@ptpcp2.phy.tu-dresden.de>
	(Arnd Baecker's message of "Tue, 4 May 2004 14:49:33 +0200 (CEST)")
References: <07C6A61102C94148B8104D42DE95F7E86DECEA@exchange2k.envision.co.il>
	<Pine.LNX.4.51.0405041440130.12522@ptpcp2.phy.tu-dresden.de>
Message-ID: <m2isfcco8k.fsf@mother.paradise.lost>

>>>>> "Arnd" == Arnd Baecker <arnd.baecker at web.de> writes:

    Arnd> There is a patch on sourceforge for gnuplot which allows for
    Arnd> bitmap images (ie. matrices). I once tried it out and it is
    Arnd> very nice.  It might take a little time until this gets
    Arnd> integrated as the gnuplot team seems to be a bit busy with
    Arnd> after release issues.

    Arnd> If you really want to go for large matrices have a look at
    Arnd> MayaVi, http://mayavi.sourceforge.net/ In particular
    Arnd> http://mayavi.sourceforge.net/docs/guide/x967.html#TOOLS

Also the fairly new imshow in matplotlib plots grayscale and
pseudocolor numeric/numarray arrays.  This replaces the painfully slow
pcolor which was unusable except for small arrays.  The guts of imshow
are implemented entirely in numeric and extension code and should
handle large arrays efficiently.  If you encounter performance issues,
I'd like to know about it!

Right now the only false colormap is ColormapJet (see
examples/pcolor_demo2.py) but you can define your own by subclassing
matplotlib.colors.Colormap.

JDH



From eric at enthought.com  Tue May  4 11:03:18 2004
From: eric at enthought.com (eric jones)
Date: Tue, 04 May 2004 10:03:18 -0500
Subject: [SciPy-user] Help with array functions
In-Reply-To: <9B40DB5FBF7D2048AD79C2A91F862A51772FFE@STEVAPPS05.Stevenage.CGI-Europe.com>
References: <9B40DB5FBF7D2048AD79C2A91F862A51772FFE@STEVAPPS05.Stevenage.CGI-Europe.com>
Message-ID: <4097B0B6.50006@enthought.com>

Bob.Cowdery at CGI-Europe.com wrote:

> Hi all,
>  
> Can anyone help me convert this loop to an array function. I can't 
> have loops in my signal processing as they take far too much 
> processing power. This is a fragment from a speech processor using 
> Ferkins formula. Essentially the gain array (m_procGain) which is 2048 
> long (m_sampl_sz) needs its nth element processing into it's (n-1)th 
> element. This is eventually used to multiply the samples in 
> m_complexout, the first 2048 elements of this are used. Attack is a 
> scaler currently 0.4.
>
>                 peak = 
> abs(self.m_complexout[argmax(abs(self.m_complexout[:self.m_sampl_sz]))])
>                 i = arange(1,self.m_sampl_sz)
>                 for x in i:
>                     if abs(self.m_complexout[x-1]) > 0:
>                        self.m_procGain[x] = ( (self.m_procGain[x-1] * 
> (1 - attack)) + ((attack * peak) / abs(self.m_complexout[x-1])) )        
>                        if self.m_procGain[x] > level:
>                            self.m_procGain[x] = level
>                     else:
>                        self.m_procGain[x] = 1
>
The general ideas of uisng Numeric in Python is to avoid loops, even at 
the expense of sometimes doing extra computations.  In this case, the 
signal processing calculations can be done for every entry in the array 
(except the first).  The conditions checked for in the "if" statements 
can then be filtered out in subsequent array processing steps.


Here is some (totally untested) code that should give you some ideas:

    gain = self.m_procGain
    abs_complexout = abs(self.m_complexout)

    # Do singal processing calculation for all elements in array
    gain[1:] = ( (gain[:-1] * (1 - attack)) +
                 ((attack * peak) / abs_complexout[:-1]) )

    # clip the upper values to some level.
    gain = choose(gain > level, (gain, level))

    # if complexout is less than 0, set gain to 1.
    gain = choose(abs_complexout < 0, (gain, 1))

    self.m_procGain = gain

With numarray I think it would look something like this (some numarray 
user correct me if I'm wrong):

    gain = self.m_procGain
    abs_complexout = abs(self.m_complexout)

    # Do singal processing calculation for all elements in array
    gain[1:] = ( (gain[:-1] * (1 - attack)) +
                 ((attack * peak) / abs_complexout[:-1]) )

    # clip the upper values to some level.
    gain[gain > level] = level

    # if complexout is less than 0, set gain to 1.
    gain[abs_complexout < 0] = 1

    # unnecessary because it is a reference to the same array and all 
operations are done using
    # indexing on the LHS.
    self.m_procGain = gain

The numarray code is easier to read.  It also should be more efficient 
(if not now, then eventually).  I really like the arrays as indices to 
arrays...

eric

> Thanks for any assistance.
>  
> 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 perry at stsci.edu  Tue May  4 11:16:34 2004
From: perry at stsci.edu (Perry Greenfield)
Date: Tue, 4 May 2004 11:16:34 -0400
Subject: [SciPy-user] Help with array functions
In-Reply-To: <4097B0B6.50006@enthought.com>
Message-ID: <CEELJPECNGEGKFNDLHFFAEFMCDAA.perry@stsci.edu>

Eric Jones wrote:
> Bob.Cowdery at CGI-Europe.com wrote:
> 
> The general ideas of uisng Numeric in Python is to avoid loops, even at 
> the expense of sometimes doing extra computations.  In this case, the 
> signal processing calculations can be done for every entry in the array 
> (except the first).  The conditions checked for in the "if" statements 
> can then be filtered out in subsequent array processing steps.
> 
> 
> Here is some (totally untested) code that should give you some ideas:
> 
>     gain = self.m_procGain
>     abs_complexout = abs(self.m_complexout)
> 
>     # Do singal processing calculation for all elements in array
>     gain[1:] = ( (gain[:-1] * (1 - attack)) +
>                  ((attack * peak) / abs_complexout[:-1]) )
> 
Does this do what is needed? I think the point he was making
was that gain[i] depends on gain[i-1] which in turn depends
on gain[i-2] and so on. The expression above doesn't capture that
since from his code he doesn't show any initial value for gain
(implicitly the gain[0] must exist). So using an array expression
which uses a shifted gain array in its input won't work. 

This is the sort of problem that doesn't fit the array model
very well unless there are some primitive array operations
that happen to do the iterative thing needed (sometimes
the accumulate or convolve functions can be used for this
thing, but I can't see any obvious existing function. Seems
like a candidate for a C extension but maybe I'm missing 
something.

Perry
> 



From Bob.Cowdery at CGI-Europe.com  Tue May  4 11:36:16 2004
From: Bob.Cowdery at CGI-Europe.com (Bob.Cowdery at CGI-Europe.com)
Date: Tue, 4 May 2004 16:36:16 +0100 
Subject: [SciPy-user] Help with array functions
Message-ID: <9B40DB5FBF7D2048AD79C2A91F862A51773001@STEVAPPS05.Stevenage.CGI-Europe.com>

Eric Jones wrote:
> Bob.Cowdery at CGI-Europe.com wrote:
> 
> The general ideas of uisng Numeric in Python is to avoid loops, even 
> at
> the expense of sometimes doing extra computations.  In this case, the 
> signal processing calculations can be done for every entry in the array 
> (except the first).  The conditions checked for in the "if" statements 
> can then be filtered out in subsequent array processing steps.
> 
> 
> Here is some (totally untested) code that should give you some ideas:
> 
>     gain = self.m_procGain
>     abs_complexout = abs(self.m_complexout)
> 
>     # Do singal processing calculation for all elements in array
>     gain[1:] = ( (gain[:-1] * (1 - attack)) +
>                  ((attack * peak) / abs_complexout[:-1]) )
>

Perry Greenfield wrote: 
> Does this do what is needed? I think the point he was making was that
gain[i] depends on gain[i-1] which in turn depends on gain[i-2] 
> and so on. The expression above doesn't capture that since from his code
he doesn't show any initial value for gain (implicitly the 
> gain[0] must exist). So using an array expression which uses a shifted
gain array in its input won't work. 

> This is the sort of problem that doesn't fit the array model very well
unless there are some primitive array operations that happen to > do the
iterative thing needed (sometimes the accumulate or convolve functions can
be used for this thing, but I can't see any obvious 
> existing function. Seems like a candidate for a C extension but maybe I'm
missing 
> something.

That is correct, my basic problem was how to use the same array in input and
output which I guess I should have mentioned. The gain array is initialised
to zeros.

Bob
_______________________________________________
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/20040504/da287b2f/attachment.html>

From eric at enthought.com  Tue May  4 11:39:45 2004
From: eric at enthought.com (eric jones)
Date: Tue, 04 May 2004 10:39:45 -0500
Subject: [SciPy-user] Help with array functions
In-Reply-To: <CEELJPECNGEGKFNDLHFFAEFMCDAA.perry@stsci.edu>
References: <CEELJPECNGEGKFNDLHFFAEFMCDAA.perry@stsci.edu>
Message-ID: <4097B941.1060809@enthought.com>

Yikes.  Your right!

eric

Perry Greenfield wrote:

>Eric Jones wrote:
>  
>
>>Bob.Cowdery at CGI-Europe.com wrote:
>>
>>The general ideas of uisng Numeric in Python is to avoid loops, even at 
>>the expense of sometimes doing extra computations.  In this case, the 
>>signal processing calculations can be done for every entry in the array 
>>(except the first).  The conditions checked for in the "if" statements 
>>can then be filtered out in subsequent array processing steps.
>>
>>
>>Here is some (totally untested) code that should give you some ideas:
>>
>>    gain = self.m_procGain
>>    abs_complexout = abs(self.m_complexout)
>>
>>    # Do singal processing calculation for all elements in array
>>    gain[1:] = ( (gain[:-1] * (1 - attack)) +
>>                 ((attack * peak) / abs_complexout[:-1]) )
>>
>>    
>>
>Does this do what is needed? I think the point he was making
>was that gain[i] depends on gain[i-1] which in turn depends
>on gain[i-2] and so on. The expression above doesn't capture that
>since from his code he doesn't show any initial value for gain
>(implicitly the gain[0] must exist). So using an array expression
>which uses a shifted gain array in its input won't work. 
>
>This is the sort of problem that doesn't fit the array model
>very well unless there are some primitive array operations
>that happen to do the iterative thing needed (sometimes
>the accumulate or convolve functions can be used for this
>thing, but I can't see any obvious existing function. Seems
>like a candidate for a C extension but maybe I'm missing 
>something.
>
>Perry
>  
>
>
>_______________________________________________
>SciPy-user mailing list
>SciPy-user at scipy.net
>http://www.scipy.net/mailman/listinfo/scipy-user
>  
>




From karthik at james.hut.fi  Tue May  4 12:19:55 2004
From: karthik at james.hut.fi (Karthikesh Raju)
Date: Tue, 4 May 2004 19:19:55 +0300
Subject: [SciPy-user] concatenation of arrays woes !!
In-Reply-To: <005b01c431db$6a244fe0$01fd5644@playroom>
References: <Pine.SGI.4.58.0405041055040.5536421@james.hut.fi>
	<005b01c431db$6a244fe0$01fd5644@playroom>
Message-ID: <Pine.SGI.4.58.0405041856360.5736153@james.hut.fi>

Hi Gary and others,

thankx for your reply. i had managed to break the procedure into bits and
pieces and get the work done. i have attached the file. Is there some way
by which we can efficiently do something like

s2 = s[0:CL-d[k,l],k]
s3 = concatenate(z1,z2,s2[:,NewAxis])
stemp[:,NewAxis] = s3

is there any means by which i can avoid having s2 as a variable, and by
using the NewAxis during striding process itself.

i have just built this matrix now, row by row :), any way to improve this!

With warm regards

karthik


-----------------------------------------------------------------------
Karthikesh Raju,		    email: karthik at james.hut.fi
Researcher,			    http://www.cis.hut.fi/karthik
Helsinki University of Technology,  Tel: +358-9-451 5389
Laboratory of Comp. & Info. Sc.,    Fax: +358-9-451 3277
Department of Computer Sc.,
P.O Box 5400, FIN 02015 HUT,
Espoo, FINLAND
-----------------------------------------------------------------------

On Tue, 4 May 2004, Gary Pajer wrote:

> ----- Original Message -----
> From: "Karthikesh Raju" <karthik at james.hut.fi>
> To: <scipy-user at scipy.net>
> Sent: Tuesday, May 04, 2004 4:10 AM
> Subject: [SciPy-user] concatenation of arrays woes !!
>
>
> >
> > Hi All,
> >
> > is it possible to concatenate arrays along the column,
> >
> > A[:,1:5] = concatenate(a,b,c)
> > a is 2x1, b 2x1, c 1x1
> >
> > i have been trying to convert the following matlab code and this is my
> > stumbling block for several months :(
> >
> > K = 3
> > CL = 5
> > L = 2
> > d -- (K,L) in the range (0,CL)
> > codes -- (K,CL)
> > window = 1
> > s = codes'
> >
> > output S -(CL*window,K*L*(window+1)) matrix
> >
> > for k = 1:K
> > for l = 1:L
> >     S(:,1+(window+1)*(l-1)+(window+1)*L*(k-1)) =
> >            [s(CL-d(k,l)+1:CL,k);
> zeros(CL*(window-1),1);zeros(CL-d(k,l),1)]
> >     S(:,(window+1)*1+(window+1)*L*(k-1)) =
> >            [zeros(d(k,l),1);zeros(CL*(window-1),1);s(1:CL-d(k,l),k)]
> >     for n=2:window
> >         S(:,n+(window+1)*(l-1)+(numbits+1)*L*(k-1)) =
> >            [zeros(d(k,l),1); zeros(CL*(n-2),1);
> > s(:,k);zeros((window-n)*CL+CL-d(k,l),1)]
> >     end
> > end
> > end
> >
> > --------------------
> >
> > This module has been my stumbling block for the last 2 months, had no
> > problems with matlab, here i have all kinds of concatination woes. Please
> > give me pointers on how to achieve such concatenations,
> >
> > thanks a lot
> >
> > regards
> > karthik
>
>
>
> I'm still not sure what you are trying to achieve.  If you post a minimal
> contrived example of what you are trying to do instead of a clip from the
> middle of your code it would be easier to figure out.
>
> But have you explored r_[ ]  and c_[ ] ?   ('c' for column, 'r' for row)
> These constructions are designed to aid concatenation:
>
> >>> from scipy import *
> >>> x=array([1,2])
> >>> y=array([3,4])
>
> >>> c_[x,y]
> array([1, 2, 3, 4])
>
> >>> c_[x[:,NewAxis],y[:,NewAxis]]
> array([[1, 3],
>        [2, 4]])
>
> >>> r_[x[:,NewAxis],y[:,NewAxis]]
> array([[1],
>        [2],
>        [3],
>        [4]])
>
>
> hth,
> gary
>
> _______________________________________________
> SciPy-user mailing list
> SciPy-user at scipy.net
> http://www.scipy.net/mailman/listinfo/scipy-user
>
>
-------------- next part --------------
from numarray import *

K = 3
L = 2
CL = 5

codes = arange(1,16,shape=(K,CL))
codes.transpose()
s = array(codes, copy=1)
print s
d = ones((K,L))
window = 1

S = zeros([CL*window,K*L*(window+1)])

for k in range(K):
	for l in range(L):
		print "--"*30 + "\n"
		
		# first row
		#--------------------------------------------------
		stemp = S[:,(window+1)*(l)+(window+1)*L*(k)] 
		s2 =  s[CL-d[k,l]:CL,k]
		z1 =  zeros([CL*(window-1),1])
		z2 =  zeros([CL-d[k,l],1])
		if CL*(window-1) == 0:
		 	s3 = concatenate((s2[:,NewAxis],z2))
		else:
		 	s3 = concatenate((s2[:,NewAxis],z1,z2))	
		stemp[:,NewAxis] = s3
		#-------------------------------------------------- 


		# # second row
		#-------------------------------------------------- 
		stemp = S[:,(window+1)*l+window+(window+1)*L*k]
		z1 = zeros([d[k,l],1]);
		z2 = zeros([CL*(window-1),1])
		s2 = s[0:CL-d[k,l],k]
		if CL*(window-1) == 0:
			s3 = concatenate((z1,s2[:,NewAxis]))
		else:
			s3 = concatenate((z1,z2,s2[:,NewAxis]))
		stemp[:,NewAxis] = s3
		#-------------------------------------------------- 

		# final push
		for w in range(0,window-1):
			stemp = S[:,w+1+(window+1)*l+(window+1)*L*k]
			z1 = zeros([d[k,l],1])
			z2 = zeros([CL*w,1])
			s2 = s[:,k]
			z3 = zeros([(window-(w+2))*CL+CL-d[k,l],1])
			if CL*w == 0:
				s3 = concatenate((z1,s2[:,NewAxis],z3))
			else:
				s3 = concatenate((z1,z2,s2[:,NewAxis],z3))
			print  s3
			stemp[:,NewAxis] = s3
print S
print S.shape
		

		
		
		


From eric at enthought.com  Tue May  4 12:31:52 2004
From: eric at enthought.com (eric jones)
Date: Tue, 04 May 2004 11:31:52 -0500
Subject: [SciPy-user] Help with array functions
In-Reply-To: <4097B941.1060809@enthought.com>
References: <CEELJPECNGEGKFNDLHFFAEFMCDAA.perry@stsci.edu>
	<4097B941.1060809@enthought.com>
Message-ID: <4097C578.1090207@enthought.com>

So, try two...

I don't know of a pure Python solution either.  My next option is 
usually playing around in weave.  Here is a weave snippet that 
implements the algorithm in C.  It runs, but just like the last one is 
untested.  Given the track record here, a decent test is advisable... :-) 

Here is the output from my cursory check to see if the code would work:

C:\temp>wv.py
weave results:  [ 0. 0.8 0.96 0.992 0.9984  0.99968  0.999936  
0.9999872  0.99999744  1.]
python results: [ 0. 0.8 0.96 0.992 0.9984  0.99968  0.999936  
0.9999872  0.99999744  1.]
C time: 0.00707567096418
Python time: 0.316835262617
weave vs. Python speedup: 44.7781226997

This is using gcc as the compiler in weave and 2048 samples in a 
signal.  If you drop to 256, the improvement is cut in half.

Now the code:

# test.py
from scipy import zeros, ones, Float64, arange
from scipy import stats
import weave

def python_algorithm(gain, complexout, attack, peak, level, samples):
    # gain is changed *inplace*
    i = arange(1,samples)
    for x in i:
        if abs(complexout[x-1]) > 0:
           gain[x] = ( (gain[x-1] * (1.0 - attack)) + ((attack * peak) / 
abs(complexout[x-1])) )       
           if gain[x] > level:
               gain[x] = level
        else:
           gain[x] = 1

def weave_algorithm(gain, complexout, attack, peak, level, samples):
    # gain is changed *inplace*
    code = """
           for (int x=1; x < samples; x++)
           {
               if (abs(complexout[x-1]) > 0.0)
               {
                  gain[x] = ( (gain[x-1] * (1.0 - attack)) +
                              ((attack * peak) / abs(complexout[x-1])) 
);       
                  if (gain[x] > level)
                      gain[x] = level;
               }
               else          
                  gain[x] = 1;
           }
           """
   
    weave.inline(code,
                 ['gain', 'complexout', 'samples', 'attack', 'peak', 
'level'],
                 compiler='gcc')
   

# some default parameters...
samples = 2048
gain = zeros(samples,typecode=Float64)
complexout = stats.choice([-1.0,0.0,1.0],size=samples)

attack = .8
peak = 1.0
level = 1.0


# equivalence tests
weave_algorithm(gain, complexout, attack, peak, level, samples)
print 'weave results:', gain[:10]

gain = zeros(samples,typecode=Float64)
python_algorithm(gain, complexout, attack, peak, level, samples)
print 'python results:', gain[:10]


# speed tests
import time

N = 100

t1 = time.clock()
for it in range(N):
    weave_algorithm(gain, complexout, attack, peak, level, samples)
t2 = time.clock()
ct = t2 - t1
print 'C time:', ct

t1 = time.clock()
for it in range(N):
    python_algorithm(gain, complexout, attack, peak, level, samples)
t2 = time.clock()
pt = t2-t1
print 'Python time:', pt

ratio = pt/ct
print 'weave vs. Python speedup:', ratio


eric jones wrote:

> Yikes.  Your right!
>
> eric
>
> Perry Greenfield wrote:
>
>> Eric Jones wrote:
>>  
>>
>>> Bob.Cowdery at CGI-Europe.com wrote:
>>>
>>> The general ideas of uisng Numeric in Python is to avoid loops, even 
>>> at the expense of sometimes doing extra computations.  In this case, 
>>> the signal processing calculations can be done for every entry in 
>>> the array (except the first).  The conditions checked for in the 
>>> "if" statements can then be filtered out in subsequent array 
>>> processing steps.
>>>
>>>
>>> Here is some (totally untested) code that should give you some ideas:
>>>
>>>    gain = self.m_procGain
>>>    abs_complexout = abs(self.m_complexout)
>>>
>>>    # Do singal processing calculation for all elements in array
>>>    gain[1:] = ( (gain[:-1] * (1 - attack)) +
>>>                 ((attack * peak) / abs_complexout[:-1]) )
>>>
>>>   
>>
>> Does this do what is needed? I think the point he was making
>> was that gain[i] depends on gain[i-1] which in turn depends
>> on gain[i-2] and so on. The expression above doesn't capture that
>> since from his code he doesn't show any initial value for gain
>> (implicitly the gain[0] must exist). So using an array expression
>> which uses a shifted gain array in its input won't work.
>> This is the sort of problem that doesn't fit the array model
>> very well unless there are some primitive array operations
>> that happen to do the iterative thing needed (sometimes
>> the accumulate or convolve functions can be used for this
>> thing, but I can't see any obvious existing function. Seems
>> like a candidate for a C extension but maybe I'm missing something.
>>
>> Perry
>>  
>>
>>
>> _______________________________________________
>> SciPy-user mailing list
>> SciPy-user at scipy.net
>> http://www.scipy.net/mailman/listinfo/scipy-user
>>  
>>
>
>
> _______________________________________________
> SciPy-user mailing list
> SciPy-user at scipy.net
> http://www.scipy.net/mailman/listinfo/scipy-user





From oliphant at ee.byu.edu  Tue May  4 14:47:43 2004
From: oliphant at ee.byu.edu (Travis Oliphant)
Date: Tue, 04 May 2004 12:47:43 -0600
Subject: ***[Possible UCE]*** [SciPy-user] Help with array functions
In-Reply-To: <9B40DB5FBF7D2048AD79C2A91F862A51772FFE@STEVAPPS05.Stevenage.CGI-Europe.com>
References: <9B40DB5FBF7D2048AD79C2A91F862A51772FFE@STEVAPPS05.Stevenage.CGI-Europe.com>
Message-ID: <4097E54F.9020007@ee.byu.edu>

Bob.Cowdery at CGI-Europe.com wrote:

> Hi all,
>  
> Can anyone help me convert this loop to an array function. I can't 
> have loops in my signal processing as they take far too much 
> processing power. This is a fragment from a speech processor using 
> Ferkins formula. Essentially the gain array (m_procGain) which is 2048 
> long (m_sampl_sz) needs its nth element processing into it's (n-1)th 
> element. This is eventually used to multiply the samples in 
> m_complexout, the first 2048 elements of this are used. Attack is a 
> scaler currently 0.4.
>
>                 peak = 
> abs(self.m_complexout[argmax(abs(self.m_complexout[:self.m_sampl_sz]))])
>                 i = arange(1,self.m_sampl_sz)
>                 for x in i:
>                     if abs(self.m_complexout[x-1]) > 0:
>                        self.m_procGain[x] = ( (self.m_procGain[x-1] * 
> (1 - attack)) + ((attack * peak) / abs(self.m_complexout[x-1])) )        
>                        if self.m_procGain[x] > level:
>                            self.m_procGain[x] = level
>                     else:
>                        self.m_procGain[x] = 1
>

Bob,

You have here something similar to an IIR filter.  The command 
signal.lfilter is usually used in situations like this.

If I read this code correctly then define y[n] = self.m_procGain[n]  
with a = (1-attack)

w[n] = (attack*peak) / abs(self.m_complexout[n])

and you want to basically compute

y[n] = a y[n-1] + w[n-1]  

as long as w[n-1] is not infinite because of zero-valued 
abs(self.m_complexout[n-1])

The problem with using signal.lfilter is you have a non-linear filter 
due to the if statements.  You could try to modify the algorithm a bit 
to maintain a linear filter, or you will have to use weave (or f2py). 

Just to expose you to the many ways to accomplish the problem.  Here is 
another test using f2py



******** Results *********************
weave results: [ 0.      0.8     0.96    0.992   0.9984  1.      1.      
1.      1.   
       1.    ]
python results: [ 0.      0.8     0.96    0.992   0.9984  1.      
1.      1.      1.   
       1.    ]
f2py results: [ 0.      0.8     0.96    0.992   0.9984  1.      1.      
1.      1.   
       1.    ]
C time: 0.08
Python time: 6.13
f2py time: 0.05
weave vs. Python speedup: 76.625
f2py vs. Python speedup: 122.6


********* Code below ***********



# test.py

fort_code = r"""

       subroutine gain_compute(gain, abs_complexout, attack, peak,
     $        level, samples)

       integer samples
       double precision attack, peak, level
       double precision gain(samples), abs_complexout(samples)

       double precision val
       integer i

       do i = 2, samples
           if (abs_complexout(i-1).gt.0.0) then
               val = gain(i-1)*(1.0-attack) +
     $                 (attack*peak) / abs_complexout(i-1)
               if (val.gt.level) then
                   val = level
               end if
           else
              val = 1.0
           end if
           gain(i) = val
       end do

       return
       end subroutine gain_compute
"""
     

from scipy import zeros, ones, Float64, arange
from scipy import stats
import scipy_distutils
import weave
import f2py2e

try:
    import forttest
except ImportError:
    f2py2e.compile(fort_code, modulename='forttest')
    import forttest

def python_algorithm(gain, abs_complexout, attack, peak, level, samples):
   # gain is changed *inplace*
   i = arange(1,samples)
   for x in i:
       if abs_complexout[x-1] > 0:
          gain[x] = ( (gain[x-1] * (1.0 - attack)) + ((attack * peak) / 
abs_complexout[x-1]) )
          if gain[x] > level:
              gain[x] = level
       else:
          gain[x] = 1

def weave_algorithm(gain, abs_complexout, attack, peak, level, samples):
   # gain is changed *inplace*
   code = """
          for (int x=1; x < samples; x++)
          {
              if (abs_complexout[x-1] > 0.0)
              {
                 gain[x] = ( (gain[x-1] * (1.0 - attack)) +
                             ((attack * peak) / abs_complexout[x-1]) 
);                        if (gain[x] > level)
                     gain[x] = level;
              }
              else
                 gain[x] = 1;
          }
          """
   weave.inline(code,
                ['gain', 'abs_complexout', 'samples', 'attack', 'peak', 
'level'],
                compiler='gcc')
 
# some default parameters...
samples = 2048
gain = zeros(samples,typecode=Float64)
complexout = stats.choice([-1.0,0.0,1.0],size=samples)

attack = .8
peak = 1.0
level = 1.0


# equivalence tests
ac = abs(complexout)
weave_algorithm(gain, ac, attack, peak, level, samples)
print 'weave results:', gain[:10]

gain = zeros(samples,typecode=Float64)
python_algorithm(gain, ac, attack, peak, level, samples)
print 'python results:', gain[:10]

gain = zeros(samples,typecode=Float64)
forttest.gain_compute(gain, ac, attack, peak, level, samples)
print 'f2py results:', gain[:10]


# speed tests
import time

N = 1000

t1 = time.clock()
for it in range(N):
   weave_algorithm(gain, complexout, attack, peak, level, samples)
t2 = time.clock()
ct = t2 - t1
print 'C time:', ct

t1 = time.clock()
for it in range(N):
   python_algorithm(gain, complexout, attack, peak, level, samples)
t2 = time.clock()
pt = t2-t1
print 'Python time:', pt

f2py_algorithm = forttest.gain_compute
t1 = time.clock()
for it in range(N):
   f2py_algorithm(gain, complexout, attack, peak, level, samples)
t2 = time.clock()
ft = t2-t1
print 'f2py time:', ft


ratio = pt/ct
ratio2 = pt/ft
print 'weave vs. Python speedup:', ratio
print 'f2py vs. Python speedup:', ratio2












> Thanks for any assistance.
>  
> 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 eric at enthought.com  Tue May  4 16:25:34 2004
From: eric at enthought.com (eric jones)
Date: Tue, 04 May 2004 15:25:34 -0500
Subject: ***[Possible UCE]*** [SciPy-user] Help with array functions
In-Reply-To: <4097E54F.9020007@ee.byu.edu>
References: <9B40DB5FBF7D2048AD79C2A91F862A51772FFE@STEVAPPS05.Stevenage.CGI-Europe.com>
	<4097E54F.9020007@ee.byu.edu>
Message-ID: <4097FC3E.40602@enthought.com>

Hey Travis,

The Fortran version didn't work for me.  It looks like g77 is detected 
on my machine, and I get compile errors in the code.  I also have an f90 
compiler on my machine that wasn't picked up.  Is there a way to switch 
which compiler is used through command line?

eric


Travis Oliphant wrote:

> Bob.Cowdery at CGI-Europe.com wrote:
>
>> Hi all,
>>  
>> Can anyone help me convert this loop to an array function. I can't 
>> have loops in my signal processing as they take far too much 
>> processing power. This is a fragment from a speech processor using 
>> Ferkins formula. Essentially the gain array (m_procGain) which is 
>> 2048 long (m_sampl_sz) needs its nth element processing into it's 
>> (n-1)th element. This is eventually used to multiply the samples in 
>> m_complexout, the first 2048 elements of this are used. Attack is a 
>> scaler currently 0.4.
>>
>>                 peak = 
>> abs(self.m_complexout[argmax(abs(self.m_complexout[:self.m_sampl_sz]))])
>>                 i = arange(1,self.m_sampl_sz)
>>                 for x in i:
>>                     if abs(self.m_complexout[x-1]) > 0:
>>                        self.m_procGain[x] = ( (self.m_procGain[x-1] * 
>> (1 - attack)) + ((attack * peak) / abs(self.m_complexout[x-1])) 
>> )                               if self.m_procGain[x] > level:
>>                            self.m_procGain[x] = level
>>                     else:
>>                        self.m_procGain[x] = 1
>>
>
> Bob,
>
> You have here something similar to an IIR filter.  The command 
> signal.lfilter is usually used in situations like this.
>
> If I read this code correctly then define y[n] = self.m_procGain[n]  
> with a = (1-attack)
>
> w[n] = (attack*peak) / abs(self.m_complexout[n])
>
> and you want to basically compute
>
> y[n] = a y[n-1] + w[n-1] 
> as long as w[n-1] is not infinite because of zero-valued 
> abs(self.m_complexout[n-1])
>
> The problem with using signal.lfilter is you have a non-linear filter 
> due to the if statements.  You could try to modify the algorithm a bit 
> to maintain a linear filter, or you will have to use weave (or f2py).
> Just to expose you to the many ways to accomplish the problem.  Here 
> is another test using f2py
>
>
>
> ******** Results *********************
> weave results: [ 0.      0.8     0.96    0.992   0.9984  1.      
> 1.      1.      1.         1.    ]
> python results: [ 0.      0.8     0.96    0.992   0.9984  1.      
> 1.      1.      1.         1.    ]
> f2py results: [ 0.      0.8     0.96    0.992   0.9984  1.      
> 1.      1.      1.         1.    ]
> C time: 0.08
> Python time: 6.13
> f2py time: 0.05
> weave vs. Python speedup: 76.625
> f2py vs. Python speedup: 122.6
>
>
> ********* Code below ***********
>
>
>
> # test.py
>
> fort_code = r"""
>
>       subroutine gain_compute(gain, abs_complexout, attack, peak,
>     $        level, samples)
>
>       integer samples
>       double precision attack, peak, level
>       double precision gain(samples), abs_complexout(samples)
>
>       double precision val
>       integer i
>
>       do i = 2, samples
>           if (abs_complexout(i-1).gt.0.0) then
>               val = gain(i-1)*(1.0-attack) +
>     $                 (attack*peak) / abs_complexout(i-1)
>               if (val.gt.level) then
>                   val = level
>               end if
>           else
>              val = 1.0
>           end if
>           gain(i) = val
>       end do
>
>       return
>       end subroutine gain_compute
> """
>    
> from scipy import zeros, ones, Float64, arange
> from scipy import stats
> import scipy_distutils
> import weave
> import f2py2e
>
> try:
>    import forttest
> except ImportError:
>    f2py2e.compile(fort_code, modulename='forttest')
>    import forttest
>
> def python_algorithm(gain, abs_complexout, attack, peak, level, samples):
>   # gain is changed *inplace*
>   i = arange(1,samples)
>   for x in i:
>       if abs_complexout[x-1] > 0:
>          gain[x] = ( (gain[x-1] * (1.0 - attack)) + ((attack * peak) / 
> abs_complexout[x-1]) )
>          if gain[x] > level:
>              gain[x] = level
>       else:
>          gain[x] = 1
>
> def weave_algorithm(gain, abs_complexout, attack, peak, level, samples):
>   # gain is changed *inplace*
>   code = """
>          for (int x=1; x < samples; x++)
>          {
>              if (abs_complexout[x-1] > 0.0)
>              {
>                 gain[x] = ( (gain[x-1] * (1.0 - attack)) +
>                             ((attack * peak) / abs_complexout[x-1]) 
> );                        if (gain[x] > level)
>                     gain[x] = level;
>              }
>              else
>                 gain[x] = 1;
>          }
>          """
>   weave.inline(code,
>                ['gain', 'abs_complexout', 'samples', 'attack', 'peak', 
> 'level'],
>                compiler='gcc')
>
> # some default parameters...
> samples = 2048
> gain = zeros(samples,typecode=Float64)
> complexout = stats.choice([-1.0,0.0,1.0],size=samples)
>
> attack = .8
> peak = 1.0
> level = 1.0
>
>
> # equivalence tests
> ac = abs(complexout)
> weave_algorithm(gain, ac, attack, peak, level, samples)
> print 'weave results:', gain[:10]
>
> gain = zeros(samples,typecode=Float64)
> python_algorithm(gain, ac, attack, peak, level, samples)
> print 'python results:', gain[:10]
>
> gain = zeros(samples,typecode=Float64)
> forttest.gain_compute(gain, ac, attack, peak, level, samples)
> print 'f2py results:', gain[:10]
>
>
> # speed tests
> import time
>
> N = 1000
>
> t1 = time.clock()
> for it in range(N):
>   weave_algorithm(gain, complexout, attack, peak, level, samples)
> t2 = time.clock()
> ct = t2 - t1
> print 'C time:', ct
>
> t1 = time.clock()
> for it in range(N):
>   python_algorithm(gain, complexout, attack, peak, level, samples)
> t2 = time.clock()
> pt = t2-t1
> print 'Python time:', pt
>
> f2py_algorithm = forttest.gain_compute
> t1 = time.clock()
> for it in range(N):
>   f2py_algorithm(gain, complexout, attack, peak, level, samples)
> t2 = time.clock()
> ft = t2-t1
> print 'f2py time:', ft
>
>
> ratio = pt/ct
> ratio2 = pt/ft
> print 'weave vs. Python speedup:', ratio
> print 'f2py vs. Python speedup:', ratio2
>
>
>
>
>
>
>
>
>
>
>
>
>> Thanks for any assistance.
>>  
>> 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
>>  
>>
>
> _______________________________________________
> SciPy-user mailing list
> SciPy-user at scipy.net
> http://www.scipy.net/mailman/listinfo/scipy-user





From pearu at scipy.org  Tue May  4 16:47:37 2004
From: pearu at scipy.org (Pearu Peterson)
Date: Tue, 4 May 2004 15:47:37 -0500 (CDT)
Subject: [SciPy-user] Help with array functions
In-Reply-To: <4097FC3E.40602@enthought.com>
References: <9B40DB5FBF7D2048AD79C2A91F862A51772FFE@STEVAPPS05.Stevenage.CGI-Europe.com>
	<4097E54F.9020007@ee.byu.edu> <4097FC3E.40602@enthought.com>
Message-ID: <Pine.LNX.4.58.0405041544050.2440@scipy.org>



On Tue, 4 May 2004, eric jones wrote:

> The Fortran version didn't work for me.  It looks like g77 is detected 
> on my machine, and I get compile errors in the code.  I also have an f90 
> compiler on my machine that wasn't picked up.  Is there a way to switch 
> which compiler is used through command line?

> > try:
> >    import forttest
> > except ImportError:
> >    f2py2e.compile(fort_code, modulename='forttest')

Try:

  f2py2e.compile(fort_code, modulename='forttest',
                 extra_args = '--fcompiler=gnu --compiler=mingw32')

to specify Fortran and C compilers.

Regards,
Pearu



From Bob.Cowdery at CGI-Europe.com  Wed May  5 03:54:57 2004
From: Bob.Cowdery at CGI-Europe.com (Bob.Cowdery at CGI-Europe.com)
Date: Wed, 5 May 2004 08:54:57 +0100 
Subject: [SciPy-user] Help with array functions
Message-ID: <9B40DB5FBF7D2048AD79C2A91F862A51773003@STEVAPPS05.Stevenage.CGI-Europe.com>


Pearu, Eric, Travis .. Thanks for all your help. It's going to take me a
little while to understand your replies and try them out. Real work
unfortunately keeps getting in the way! I have a similar problem with the
AGC (Automatic Gain Control) on the receive side which at the moment steps
rather than ramps as I have miscoded it along the lines of the first reply.
I will probably try to come up with a function I can use for both
situations. If you have any more inspiration in the mean time please let me
know.

Thanks
Bob

_______________________________________________
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/20040505/7656be05/attachment.html>

From Tikva at israsrv.net.il  Wed May  5 04:37:20 2004
From: Tikva at israsrv.net.il (Tikva Bonneh)
Date: Wed, 5 May 2004 11:37:20 +0300
Subject: [SciPy-user] gplt plot with errorbars 
Message-ID: <001001c4327c$323341d0$0101c80a@bonneh>

I want to plot using gplt with errorbars. In gnuplot the interface would we plot x,y,e with errorbars,
but the interface of gplt.plot is plot(x,y,'format string'), so I don't know how to pass the third parameter (the array of the error values). 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.scipy.org/pipermail/scipy-user/attachments/20040505/ba0f2814/attachment.html>

From arnd.baecker at web.de  Wed May  5 05:31:17 2004
From: arnd.baecker at web.de (Arnd Baecker)
Date: Wed, 5 May 2004 11:31:17 +0200 (CEST)
Subject: [SciPy-user] scipy on quantian
Message-ID: <Pine.LNX.4.51.0405051122020.20466@ptpcp2.phy.tu-dresden.de>

Dear SciPy users,

due to joint efforts a working SciPy is available on quantian 0.4.9.6,
"The Quantian Scientific Computing Environment",
   http://dirk.eddelbuettel.com/quantian.html
which is a live CD based on knoppix/debian.

So anyone who wants to do see what SciPy has to offer for him
can test this without installing anything.
(It might be also useful for Windows users to see
how things look like on linux).

Thanks go to Jose Fonseca for building the debian packages,
Dirk Edelbuettel for quantian and his extremely fast and helpful response
and Ulf Lorenz for a lot of testing.

Best,

Arnd

P.S.: Actually, maybe this is worth a link on the new scipy webpage?



From arnd.baecker at web.de  Wed May  5 05:31:17 2004
From: arnd.baecker at web.de (Arnd Baecker)
Date: Wed, 5 May 2004 11:31:17 +0200 (CEST)
Subject: [SciPy-user] scipy on quantian
Message-ID: <Pine.LNX.4.51.0405051122020.20466@ptpcp2.phy.tu-dresden.de>

Dear SciPy users,

due to joint efforts a working SciPy is available on quantian 0.4.9.6,
"The Quantian Scientific Computing Environment",
   http://dirk.eddelbuettel.com/quantian.html
which is a live CD based on knoppix/debian.

So anyone who wants to do see what SciPy has to offer for him
can test this without installing anything.
(It might be also useful for Windows users to see
how things look like on linux).

Thanks go to Jose Fonseca for building the debian packages,
Dirk Edelbuettel for quantian and his extremely fast and helpful response
and Ulf Lorenz for a lot of testing.

Best,

Arnd

P.S.: Actually, maybe this is worth a link on the new scipy webpage?




From chris at fonnesbeck.org  Wed May  5 10:47:23 2004
From: chris at fonnesbeck.org (Christopher Fonnesbeck)
Date: Wed, 5 May 2004 10:47:23 -0400
Subject: [SciPy-user] gplt problems: saving output to file
Message-ID: <1E7E28BC-9EA3-11D8-BA1A-000A956FDAC0@fonnesbeck.org>

I continue to have a problem saving the output from certain gplt calls  
to disk. I have two plotting routines, both which call gplt.plot, one  
which plots a histogram, and the other which plots a time series.  
Though both produce output properly to screen as expected, the latter  
produces a zero-length file. Here are the two methods:

     def  
histogram(self,data,name,nbins=None,xlab='Value',ylab='Frequency',suffix 
=''):
         'Internal histogram specification for handling nested arrays'

         'If there is only one data array, go ahead and plot it ... '
         if len(shape(data))==1:
             print 'Generating histogram of',name
             try:
                 'Generate reasonable bins if none specified'
                 if not nbins: nbins = 50 * (len(data)>1000) or 10
                 'Generate binned values'
                 counts,lowval,width,extra =  
histogram(data,numbins=nbins)
                 'Generate x-axis values'
                 values = [lowval]
                 for i in range(nbins-1):
                     values.append(values[-1]+width)
                 'Pass to plotting function'
                 plot(values,counts)
                 'Plot options'
                 grid('off')
                 xtitle(name)
                 ytitle("Frequency")
                 'Save to file'
                 output("%s%s.png" % (name,suffix),'png')
                 close()
             except OverflowError:
                 print '... cannot generate histogram'
         else:
             '... otherwise, plot recursively'
             tdata = swapaxes(data,0,1)
             for i in range(len(tdata)):
                  
self.histogram(tdata[i],name+'_'+str(i),nbins=nbins,xlab=xlab,suffix=suf 
fix)

     def time_series(self,data,name,xlab='Time',ylab='State',suffix=''):
         'Plot (mutiple) time series'

         plot(data)
         print 'Plotting',name
         grid('off')
         xtitle(xlab)
         ytitle(ylab)
         output("%s%s.png" % (name,suffix),'png')
         close()


I dont see any obvious reason why output() would produce different  
results for these two methods. If anyone has a plausible explanation  
for what may be happening, I'd love to hear from them.

Thanks,
Chris

PS -- this occurs on both Linux and OSX

--
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

"I don't remember any kind of heaviness ruining my time at Yale."
          George W. Bush, on the civil rights movement
--
Putting http://wecanstopspam.org in your email helps it pass through  
overzealous spam filters. 



From swisher at enthought.com  Wed May  5 10:49:32 2004
From: swisher at enthought.com (Janet Swisher)
Date: Wed, 5 May 2004 09:49:32 -0500
Subject: [SciPy-user] scipy on quantian
In-Reply-To: <Pine.LNX.4.51.0405051122020.20466@ptpcp2.phy.tu-dresden.de>
Message-ID: <008501c432b0$30036be0$ab01a8c0@SWISHER>


> P.S.: Actually, maybe this is worth a link on the new scipy webpage?

Done. I added a link on the Downloads page.

-------------------------
Janet Swisher
Senior Technical Writer
Enthought, Inc.
1-512-536-1057




From eric at enthought.com  Wed May  5 10:58:29 2004
From: eric at enthought.com (eric jones)
Date: Wed, 05 May 2004 09:58:29 -0500
Subject: [SciPy-user] scipy on quantian
In-Reply-To: <Pine.LNX.4.51.0405051122020.20466@ptpcp2.phy.tu-dresden.de>
References: <Pine.LNX.4.51.0405051122020.20466@ptpcp2.phy.tu-dresden.de>
Message-ID: <40990115.9080209@enthought.com>

Hey Arnd,

Very cool.

> P.S.: Actually, maybe this is worth a link on the new scipy webpage?

Perhaps we should have a link page or "related projects" page that would 
be in the "navigation" panel of the site.  I'll ask Travis V. about this.

Also, you can always post a "news" articles from your personal scipy 
page that will show up on the front page.

eric

Arnd Baecker wrote:

>Dear SciPy users,
>
>due to joint efforts a working SciPy is available on quantian 0.4.9.6,
>"The Quantian Scientific Computing Environment",
>   http://dirk.eddelbuettel.com/quantian.html
>which is a live CD based on knoppix/debian.
>
>So anyone who wants to do see what SciPy has to offer for him
>can test this without installing anything.
>(It might be also useful for Windows users to see
>how things look like on linux).
>
>Thanks go to Jose Fonseca for building the debian packages,
>Dirk Edelbuettel for quantian and his extremely fast and helpful response
>and Ulf Lorenz for a lot of testing.
>
>Best,
>
>Arnd
>
>P.S.: Actually, maybe this is worth a link on the new scipy webpage?
>
>
>_______________________________________________
>SciPy-user mailing list
>SciPy-user at scipy.net
>http://www.scipy.net/mailman/listinfo/scipy-user
>  
>




From danny_shevitz at yahoo.com  Wed May  5 11:08:42 2004
From: danny_shevitz at yahoo.com (danny shevitz)
Date: Wed, 5 May 2004 08:08:42 -0700 (PDT)
Subject: [SciPy-user] gplt plot with errorbars 
In-Reply-To: <001001c4327c$323341d0$0101c80a@bonneh>
Message-ID: <20040505150842.65035.qmail@web41008.mail.yahoo.com>

I posted this essentially same email about 3 weeks ago. The new_plot
module allegedly wraps the errorbar plot from gplot. I was never able
to get it to work. I asked for a working code sample, but none was ever
posted. I have since moved to matplotlib. It was a lot easier to get up
and running and is working well for me. There was a sample errorbar
plot in the matplotlib source file. Get that and you'll be off and
running

Danny


--- Tikva Bonneh <Tikva at israsrv.net.il> wrote:
> I want to plot using gplt with errorbars. In gnuplot the interface
> would we plot x,y,e with errorbars,
> but the interface of gplt.plot is plot(x,y,'format string'), so I
> don't know how to pass the third parameter (the array of the error
> values). 
> > _______________________________________________
> SciPy-user mailing list
> SciPy-user at scipy.net
> http://www.scipy.net/mailman/listinfo/scipy-user
> 



	
		
__________________________________
Do you Yahoo!?
Win a $20,000 Career Makeover at Yahoo! HotJobs  
http://hotjobs.sweepstakes.yahoo.com/careermakeover 



From arnd.baecker at web.de  Wed May  5 11:13:09 2004
From: arnd.baecker at web.de (Arnd Baecker)
Date: Wed, 5 May 2004 17:13:09 +0200 (CEST)
Subject: [SciPy-user] scipy on quantian
In-Reply-To: <008501c432b0$30036be0$ab01a8c0@SWISHER>
References: <008501c432b0$30036be0$ab01a8c0@SWISHER>
Message-ID: <Pine.LNX.4.51.0405051708440.20466@ptpcp2.phy.tu-dresden.de>



On Wed, 5 May 2004, Janet Swisher wrote:

> > P.S.: Actually, maybe this is worth a link on the new scipy webpage?
>
> Done. I added a link on the Downloads page.

Excellent!
Maybe it is useful to point out on that page
that there are also the debian packages by Jose Fonseca,
http://jrfonseca.dyndns.org/debian/

Many thanks,

Arnd



From arnd.baecker at web.de  Wed May  5 11:16:01 2004
From: arnd.baecker at web.de (Arnd Baecker)
Date: Wed, 5 May 2004 17:16:01 +0200 (CEST)
Subject: [SciPy-user] scipy on quantian
In-Reply-To: <40990115.9080209@enthought.com>
References: <Pine.LNX.4.51.0405051122020.20466@ptpcp2.phy.tu-dresden.de>
	<40990115.9080209@enthought.com>
Message-ID: <Pine.LNX.4.51.0405051713260.20466@ptpcp2.phy.tu-dresden.de>



On Wed, 5 May 2004, eric jones wrote:

> Hey Arnd,
>
> Very cool.
>
> > P.S.: Actually, maybe this is worth a link on the new scipy webpage?
>
> Perhaps we should have a link page or "related projects" page that would
> be in the "navigation" panel of the site.  I'll ask Travis V. about this.

I think Janet's solution to put it on http://www.scipy.org/download/
is quite good.

> Also, you can always post a "news" articles from your personal scipy
> page that will show up on the front page.

Ok, I wasn't aware of this, should I still post a news?
(If so I will look into that tomorrow).

Best,

Arnd



From arnd.baecker at web.de  Wed May  5 11:48:01 2004
From: arnd.baecker at web.de (Arnd Baecker)
Date: Wed, 5 May 2004 17:48:01 +0200 (CEST)
Subject: [SciPy-user] help and lazy importer (again or still ?)
Message-ID: <Pine.LNX.4.51.0405051746040.20466@ptpcp2.phy.tu-dresden.de>

Hi,

with yesterdays CVS version I came across the
following problem

>>> help ("scipy.integrate.odeint")
   [[ this works fine ]]
>>> from  scipy.integrate import odeint
>>> help ("scipy.integrate.odeint")
Traceback (m`ost recent call last):
  File "<stdin>", line 1, in ?
  File "/home/python/PYTHON_New//lib/python2.3/site.py", line 309, in
__call__
    return pydoc.help(*args, **kwds)
  File
"/home/python/PYTHON_New/lib/python2.3/site-packages/scipy_base/ppimport.py",
line 307, in _scipy_pydoc_help_call
    a = a._ppimport_module
  File
"/home/python/PYTHON_New/lib/python2.3/site-packages/scipy_base/ppimport.py",
line 270, in __getattr__
    module = self._ppimport_importer()
  File
"/home/python/PYTHON_New/lib/python2.3/site-packages/scipy_base/ppimport.py",
line 243, in _ppimport_importer
    module = __import__(name,None,None,['*'])
ImportError: No module named odeint
>>>

After importing (via `from  scipy.integrate import odeint`)
>>>help(odeint)
works (as expected).

Our students stumbled across this and complained
"There is no help for scipy.integrate.odeint" ;-).

I hope that there is an easy solution to this
(and that my question does not open a can of worms ... ;-)

And there is another (related?) one:

Python 2.3.3 (#1, May  3 2004, 16:38:25)
[GCC 3.3.3 (Debian)] on linux2
Type "help", "copyright", "credits" or "license" for more information.
>>> help("scipy.linalg")
Gives
  Help on instance of _ModuleLoader in scipy:
  scipy.linalg = <module 'scipy.linalg' from '/home/python/PYTHON...
    e-packages/scipy/linalg/__init__.pyc' [imported]>

Only on the second call to help("scipy.linalg")
one gets the expected documentation.

Best,

Arnd



From eric at enthought.com  Wed May  5 12:11:32 2004
From: eric at enthought.com (eric jones)
Date: Wed, 05 May 2004 11:11:32 -0500
Subject: [SciPy-user] scipy on quantian
In-Reply-To: <Pine.LNX.4.51.0405051713260.20466@ptpcp2.phy.tu-dresden.de>
References: <Pine.LNX.4.51.0405051122020.20466@ptpcp2.phy.tu-dresden.de>
	<40990115.9080209@enthought.com>
	<Pine.LNX.4.51.0405051713260.20466@ptpcp2.phy.tu-dresden.de>
Message-ID: <40991234.6020204@enthought.com>

Arnd Baecker wrote:

>On Wed, 5 May 2004, eric jones wrote:
>
>  
>
>>Hey Arnd,
>>
>>Very cool.
>>
>>    
>>
>>>P.S.: Actually, maybe this is worth a link on the new scipy webpage?
>>>      
>>>
>>Perhaps we should have a link page or "related projects" page that would
>>be in the "navigation" panel of the site.  I'll ask Travis V. about this.
>>    
>>
>
>I think Janet's solution to put it on http://www.scipy.org/download/
>is quite good.
>
>  
>
Fine with me also.

>>Also, you can always post a "news" articles from your personal scipy
>>page that will show up on the front page.
>>    
>>
>
>Ok, I wasn't aware of this, should I still post a news?
>(If so I will look into that tomorrow).
>
>  
>
I think it would be good.

eric

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




From pearu at scipy.org  Wed May  5 15:44:54 2004
From: pearu at scipy.org (Pearu Peterson)
Date: Wed, 5 May 2004 14:44:54 -0500 (CDT)
Subject: [SciPy-user] help and lazy importer (again or still ?)
In-Reply-To: <Pine.LNX.4.51.0405051746040.20466@ptpcp2.phy.tu-dresden.de>
References: <Pine.LNX.4.51.0405051746040.20466@ptpcp2.phy.tu-dresden.de>
Message-ID: <Pine.LNX.4.58.0405051430410.20584@scipy.org>



On Wed, 5 May 2004, Arnd Baecker wrote:

> with yesterdays CVS version I came across the
> following problem
> 
> >>> help ("scipy.integrate.odeint")
>    [[ this works fine ]]
> >>> from  scipy.integrate import odeint
> >>> help ("scipy.integrate.odeint")
<snip>
> ImportError: No module named odeint

> 
> Our students stumbled across this and complained
> "There is no help for scipy.integrate.odeint" ;-).
> 
> I hope that there is an easy solution to this
> (and that my question does not open a can of worms ... ;-)

Luckily;-), it wasn't a bug but an overlooked case (I believe it was 
the last one). The corresponding support code is commited to CVS now.

> And there is another (related?) one:
> 
> Python 2.3.3 (#1, May  3 2004, 16:38:25)
> [GCC 3.3.3 (Debian)] on linux2
> Type "help", "copyright", "credits" or "license" for more information.
> >>> help("scipy.linalg")
> Gives
>   Help on instance of _ModuleLoader in scipy:
>   scipy.linalg = <module 'scipy.linalg' from '/home/python/PYTHON...
>     e-packages/scipy/linalg/__init__.pyc' [imported]>
> 
> Only on the second call to help("scipy.linalg")
> one gets the expected documentation.

Unfortunately, the only way to avoid the above issue is to import 
scipy_base or scipy (that imports ppimport module which adds a hook to 
pydoc.help to handle postponed modules) before trying to get help of scipy
objects. You can import scipy_base in $PYTHONSTARTUP script, for instance.

Regards,
Pearu



From Tikva at israsrv.net.il  Wed May  5 15:41:50 2004
From: Tikva at israsrv.net.il (Tikva Bonneh)
Date: Wed, 5 May 2004 22:41:50 +0300
Subject: [SciPy-user] matplotlib versus gplt multiple windows - (errorbars
	subjet continued)
Message-ID: <004701c432d9$06261e70$0101c80a@bonneh>

I have tried matplotlib and migrated to gplt because I couldn't open multiple windows with matplotlib . Is there a way to do it?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.scipy.org/pipermail/scipy-user/attachments/20040505/f31ac6af/attachment.html>

From jdhunter at ace.bsd.uchicago.edu  Wed May  5 15:21:55 2004
From: jdhunter at ace.bsd.uchicago.edu (John Hunter)
Date: Wed, 05 May 2004 14:21:55 -0500
Subject: [SciPy-user] matplotlib versus gplt multiple windows -
	(errorbars subjet continued)
In-Reply-To: <004701c432d9$06261e70$0101c80a@bonneh> ("Tikva Bonneh"'s
	message of "Wed, 5 May 2004 22:41:50 +0300")
References: <004701c432d9$06261e70$0101c80a@bonneh>
Message-ID: <m24qquy90s.fsf@mother.paradise.lost>

>>>>> "Tikva" == Tikva Bonneh <Tikva at israsrv.net.il> writes:

    Tikva> I have tried matplotlib and migrated to gplt because I
    Tikva> couldn't open multiple windows with matplotlib . Is there a
    Tikva> way to do

Sure

from matplotlib.matlab import *

figure(1)
plot([1,2,3])

figure(2)
plot([4,5,6])

show()

See http://matplotlib.sourceforge.net/tutorial.html and the many
examples in the examples directory of the source distribution,
especially examples/multiple_figs_demo.py.

Have fun and good luck!
John Hunter 



From oliphant at ee.byu.edu  Wed May  5 15:51:41 2004
From: oliphant at ee.byu.edu (Travis E. Oliphant)
Date: Wed, 05 May 2004 13:51:41 -0600
Subject: [SciPy-user] Marginal distributions
Message-ID: <409945CD.9050600@ee.byu.edu>


Scott,

If you are trying to integrate a multidimensional function you can 
easily have an upper limit that is a variable.

You will probably want to read about

scipy.integrate.quad  for performing the actual integration.

and

scipy.vectorize (for vectorizing the resulting function).


If you could be a bit more specific about the problem you are referring 
to I might be able to help further.

My understanding of marginals density functions is that you compute 
integrals over the entire domain (no variable in the limit) so perhaps 
I'm not understanding your problem well enough.

Post more to see if we can help.

-Travis O.






From arnd.baecker at web.de  Wed May  5 15:49:25 2004
From: arnd.baecker at web.de (Arnd Baecker)
Date: Wed, 5 May 2004 21:49:25 +0200 (CEST)
Subject: [SciPy-user] gplt problems: saving output to file
In-Reply-To: <1E7E28BC-9EA3-11D8-BA1A-000A956FDAC0@fonnesbeck.org>
References: <1E7E28BC-9EA3-11D8-BA1A-000A956FDAC0@fonnesbeck.org>
Message-ID: <Pine.LNX.4.51.0405052216290.1305@ptplp09.phy.tu-dresden.de>

Hi Chris,

which gnuplot version are you using? I think
there was a time when I had a similar problem
If it's not gnuplot 4.0 you could try upgrading.
Or you could try to use the IPython+Gnuplot.py interface to
gnuplot - I am using that a lot  and without any problems.

In the histogram case: do you supply one data array
or several? (I am asking, because in the latter case
output is called more than once).

Best, Arnd


On Wed, 5 May 2004, Christopher Fonnesbeck wrote:

> I continue to have a problem saving the output from certain gplt calls
> to disk. I have two plotting routines, both which call gplt.plot, one
> which plots a histogram, and the other which plots a time series.
> Though both produce output properly to screen as expected, the latter
> produces a zero-length file. Here are the two methods:
>
>      def
> histogram(self,data,name,nbins=None,xlab='Value',ylab='Frequency',suffix
> =''):
>          'Internal histogram specification for handling nested arrays'
>
>          'If there is only one data array, go ahead and plot it ... '
>          if len(shape(data))==1:
>              print 'Generating histogram of',name
>              try:
>                  'Generate reasonable bins if none specified'
>                  if not nbins: nbins = 50 * (len(data)>1000) or 10
>                  'Generate binned values'
>                  counts,lowval,width,extra =
> histogram(data,numbins=nbins)
>                  'Generate x-axis values'
>                  values = [lowval]
>                  for i in range(nbins-1):
>                      values.append(values[-1]+width)
>                  'Pass to plotting function'
>                  plot(values,counts)
>                  'Plot options'
>                  grid('off')
>                  xtitle(name)
>                  ytitle("Frequency")
>                  'Save to file'
>                  output("%s%s.png" % (name,suffix),'png')
>                  close()
>              except OverflowError:
>                  print '... cannot generate histogram'
>          else:
>              '... otherwise, plot recursively'
>              tdata = swapaxes(data,0,1)
>              for i in range(len(tdata)):
>
> self.histogram(tdata[i],name+'_'+str(i),nbins=nbins,xlab=xlab,suffix=suf
> fix)
>
>      def time_series(self,data,name,xlab='Time',ylab='State',suffix=''):
>          'Plot (mutiple) time series'
>
>          plot(data)
>          print 'Plotting',name
>          grid('off')
>          xtitle(xlab)
>          ytitle(ylab)
>          output("%s%s.png" % (name,suffix),'png')
>          close()
>
>
> I dont see any obvious reason why output() would produce different
> results for these two methods. If anyone has a plausible explanation
> for what may be happening, I'd love to hear from them.
>
> Thanks,
> Chris
>
> PS -- this occurs on both Linux and OSX
>
> --
> 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
>
> "I don't remember any kind of heaviness ruining my time at Yale."
>           George W. Bush, on the civil rights movement
> --
> Putting http://wecanstopspam.org in your email helps it pass through
> overzealous spam filters.
>
> _______________________________________________
> SciPy-user mailing list
> SciPy-user at scipy.net
> http://www.scipy.net/mailman/listinfo/scipy-user
>



From Tikva at israsrv.net.il  Wed May  5 15:56:21 2004
From: Tikva at israsrv.net.il (Tikva Bonneh)
Date: Wed, 5 May 2004 22:56:21 +0300
Subject: [SciPy-user] matplotlib multiple windows
Message-ID: <005b01c432db$0c9ba4d0$0101c80a@bonneh>

My last pot was too short! I meant to say - I can open multiple figures in the same time, but what I wanted to do is show one figure and then open a new window while the old window still shows. I did not find how to do that with matplotlib. I had to close the first  window in order to see the next window.  
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.scipy.org/pipermail/scipy-user/attachments/20040505/931702ae/attachment.html>

From arnd.baecker at web.de  Wed May  5 17:04:53 2004
From: arnd.baecker at web.de (Arnd Baecker)
Date: Wed, 5 May 2004 23:04:53 +0200 (CEST)
Subject: [SciPy-user] help and lazy importer (again or still ?)
In-Reply-To: <Pine.LNX.4.58.0405051430410.20584@scipy.org>
References: <Pine.LNX.4.51.0405051746040.20466@ptpcp2.phy.tu-dresden.de>
	<Pine.LNX.4.58.0405051430410.20584@scipy.org>
Message-ID: <Pine.LNX.4.51.0405052326100.1305@ptplp09.phy.tu-dresden.de>

On Wed, 5 May 2004, Pearu Peterson wrote:

[...]

> Luckily;-), it wasn't a bug but an overlooked case (I believe it was
> the last one). The corresponding support code is commited to CVS now.

Great - many thanks!

> > And there is another (related?) one:
> >
> > Python 2.3.3 (#1, May  3 2004, 16:38:25)
> > [GCC 3.3.3 (Debian)] on linux2
> > Type "help", "copyright", "credits" or "license" for more information.
> > >>> help("scipy.linalg")
> > Gives
> >   Help on instance of _ModuleLoader in scipy:
> >   scipy.linalg = <module 'scipy.linalg' from '/home/python/PYTHON...
> >     e-packages/scipy/linalg/__init__.pyc' [imported]>
> >
> > Only on the second call to help("scipy.linalg")
> > one gets the expected documentation.
>
> Unfortunately, the only way to avoid the above issue is to import
> scipy_base or scipy (that imports ppimport module which adds a hook to
> pydoc.help to handle postponed modules) before trying to get help of scipy
> objects. You can import scipy_base in $PYTHONSTARTUP script, for instance.

Well, there is always a price to pay at some point ;-)
Still I am wondering why
  help("scipy.integrate.odeint")
works at a bare python prompt, whereas
  help("scipy.integrate")
doesn't. The latter seems simpler.
I am not sure if I should be able to understand this
behavior (somehow I think that in the first case it
has to look at scipy.integrate first and then at what
is in there.  The first "look" at scipy.integrate basically adds the
hook to pydoc.help so that the odeint documentation can be displayed -
but this is just guessing ;-).

Just one last question on this, and I will really shut up:
are there any implications/problems for doc-string extracting
tools (pydoc, epydoc and such)?
I'd guess not has they have to look at scipy first
and then the hook should be installed.

Best, Arnd



From jdhunter at ace.bsd.uchicago.edu  Wed May  5 16:48:51 2004
From: jdhunter at ace.bsd.uchicago.edu (John Hunter)
Date: Wed, 05 May 2004 15:48:51 -0500
Subject: [SciPy-user] matplotlib multiple windows
In-Reply-To: <005b01c432db$0c9ba4d0$0101c80a@bonneh> ("Tikva Bonneh"'s
	message of "Wed, 5 May 2004 22:56:21 +0300")
References: <005b01c432db$0c9ba4d0$0101c80a@bonneh>
Message-ID: <m2d65iwqfg.fsf@mother.paradise.lost>

>>>>> "Tikva" == Tikva Bonneh <Tikva at israsrv.net.il> writes:

    Tikva> My last pot was too short! I meant to say - I can open
    Tikva> multiple figures in the same time, but what I wanted to do
    Tikva> is show one figure and then open a new window while the old
    Tikva> window still shows. I did not find how to do that with
    Tikva> matplotlib. I had to close the first window in order to see
    Tikva> the next window.

I'm sorry, I don't really understand what it is you can and cannot do.
Could you please describe it more clearly?  Are you working
interactively from a python shell or running a script?  Are you
working in a python IDE such as idle or pycrust?  What matplotlib
backend are you using?

See also 

  http://matplotlib.sourceforge.net/interactive.html

  and

  http://matplotlib.sourceforge.net/faq.html#SHOW

Cheers,
John Hunter



From gazzar at email.com  Wed May  5 22:13:19 2004
From: gazzar at email.com (Gary Ruben)
Date: Thu, 06 May 2004 12:13:19 +1000
Subject: [SciPy-user] gplt plot with errorbars
Message-ID: <20040506021319.053441CE303@ws3-6.us4.outblaze.com>

If you want to use gnuplot, I could provide sample code for doing xyerrorbar plots via gnuplot.py if that's helpful. I can't help with gplt though.
Matplotlib is easier to get successful plots with though.
I use Matplotlib and gnuplot.py; Matplotlib for quickly plotting stuff and gnuplot.py for publication type stuff because I'm Windows based and I like the convenience of copying and pasting plots into my word processor via the clipboard.

Gary Ruben

----- Original Message -----
From: danny shevitz <danny_shevitz at yahoo.com>
Date: Wed, 5 May 2004 08:08:42 -0700 (PDT)
To: SciPy Users List <scipy-user at scipy.net>
Subject: Re: [SciPy-user] gplt plot with errorbars 

> I posted this essentially same email about 3 weeks ago. The new_plot
> module allegedly wraps the errorbar plot from gplot. I was never able
> to get it to work. I asked for a working code sample, but none was ever
> posted. I have since moved to matplotlib. It was a lot easier to get up
> and running and is working well for me. There was a sample errorbar
> plot in the matplotlib source file. Get that and you'll be off and
> running
> 
> Danny
> 
> 
> --- Tikva Bonneh <Tikva at israsrv.net.il> wrote:
> > I want to plot using gplt with errorbars. In gnuplot the interface
> > would we plot x,y,e with errorbars,
> > but the interface of gplt.plot is plot(x,y,'format string'), so I
> > don't know how to pass the third parameter (the array of the error
> > values). 

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



From pearu at scipy.org  Thu May  6 02:56:37 2004
From: pearu at scipy.org (Pearu Peterson)
Date: Thu, 6 May 2004 01:56:37 -0500 (CDT)
Subject: [SciPy-user] help and lazy importer (again or still ?)
In-Reply-To: <Pine.LNX.4.51.0405052326100.1305@ptplp09.phy.tu-dresden.de>
References: <Pine.LNX.4.51.0405051746040.20466@ptpcp2.phy.tu-dresden.de>
	<Pine.LNX.4.58.0405051430410.20584@scipy.org>
	<Pine.LNX.4.51.0405052326100.1305@ptplp09.phy.tu-dresden.de>
Message-ID: <Pine.LNX.4.58.0405060143210.20584@scipy.org>



On Wed, 5 May 2004, Arnd Baecker wrote:

> > Unfortunately, the only way to avoid the above issue is to import
> > scipy_base or scipy (that imports ppimport module which adds a hook to
> > pydoc.help to handle postponed modules) before trying to get help of scipy
> > objects. You can import scipy_base in $PYTHONSTARTUP script, for instance.
> 
> Well, there is always a price to pay at some point ;-)

It turns out I was wrong, the latest patch in CVS makes also

  help("scipy.integrate")

to work. 

> Just one last question on this, and I will really shut up:
> are there any implications/problems for doc-string extracting
> tools (pydoc, epydoc and such)?
> I'd guess not has they have to look at scipy first
> and then the hook should be installed.

The answer depends on how these tools work internally. If you find any 
problems, please, let me know. Usually the fixes are simple.

Thanks,
Pearu



From arnd.baecker at web.de  Thu May  6 03:03:33 2004
From: arnd.baecker at web.de (Arnd Baecker)
Date: Thu, 6 May 2004 09:03:33 +0200 (CEST)
Subject: [SciPy-user] help and lazy importer (again or still ?)
In-Reply-To: <Pine.LNX.4.58.0405060143210.20584@scipy.org>
References: <Pine.LNX.4.51.0405051746040.20466@ptpcp2.phy.tu-dresden.de>
	<Pine.LNX.4.58.0405051430410.20584@scipy.org>
	<Pine.LNX.4.51.0405052326100.1305@ptplp09.phy.tu-dresden.de>
	<Pine.LNX.4.58.0405060143210.20584@scipy.org>
Message-ID: <Pine.LNX.4.51.0405060859030.1494@ptpcp2.phy.tu-dresden.de>



On Thu, 6 May 2004, Pearu Peterson wrote:

>
>
> On Wed, 5 May 2004, Arnd Baecker wrote:
>
> > > Unfortunately, the only way to avoid the above issue is to import
> > > scipy_base or scipy (that imports ppimport module which adds a hook to
> > > pydoc.help to handle postponed modules) before trying to get help of scipy
> > > objects. You can import scipy_base in $PYTHONSTARTUP script, for instance.
> >
> > Well, there is always a price to pay at some point ;-)
>
> It turns out I was wrong, the latest patch in CVS makes also
>
>   help("scipy.integrate")
>
> to work.

Glad that you were wrong ;-). I think the on-line
help is an extremely helpful aspect of python/scipy,
so that's why I am very happy to see this working!
(though I think it might be nice to have some well-defined
and _simple_ procedure to add documention to scipy -
something like this is just being discussed for python on c.l.p. -
but this is off-topic here and might be good for a new thread
at some point)

_Many_ thanks,

Arnd



From nadavh at visionsense.com  Thu May  6 06:34:09 2004
From: nadavh at visionsense.com (Nadav Horesh)
Date: Thu, 6 May 2004 12:34:09 +0200
Subject: [SciPy-user] Re: xplt documentation
Message-ID: <07C6A61102C94148B8104D42DE95F7E86DECEF@exchange2k.envision.co.il>




-----Original Message-----
From:	Arnd Baecker [mailto:arnd.baecker at web.de]
Sent:	Tue 04-May-04 14:49
To:	Nadav Horesh
Cc:	scipy-user at scipy.net
Subject:	Re: [SciPy-user] Re: xplt documentation
Hi,


On Tue, 4 May 2004, Nadav Horesh wrote:

>> I've made some testing (not exhaustive) of matplotlib under IDLE and
>> pycrust on win32 and linux platforms. matplotlib seems to be more
>> tolerant then xplt,

> Can you be a bit more explicit, i.e. give examples ?
I just succeeded to interact with the plot window (zoom for instance), while in xplt I had to call pyg_pending() in order to render the mouse commands.


>> especially when show() is called instead of
>> using the interactive mode.

>Personally I think that this is one of the big strengths
>of scipy.xplt. For example, when debugging
>I use a lot
>  from IPython.Shell import IPythonShellEmbed
>  ipshell = IPythonShellEmbed()
>
>  ipshell()
>
>The ipshell() can come at any point of your code
>and all variables are available.
>So a quick
>  plg(some_array)
>or
>  pli(some_2d_array)
>often allows to find bugs much more quickly
>than anything else.

I will try that.

>> The linux environment seems to work a
>> bit better then the win32.
>>
>> There is a project (not very active nowadays) called glplot
>> (glplot.sf.net) which fills a shortcoming of gnuplot of
>> generating false colour map of large matrices (gnuplot is too slow here).

>There is a patch on sourceforge for gnuplot which allows
>for bitmap images (ie. matrices). I once tried it out and it is very nice.
>It might take a little time until this gets integrated
>as the gnuplot team seems to be a bit busy with
>after release issues.

Can you explain more about this patch --- how to access, compile and use it?
I am familiar only with the "set pm3d map" option, wich is very nice, but sill too slow for large matrices.

>If you really want to go for large matrices have a look
>at MayaVi,
>  http://mayavi.sourceforge.net/
>In particular
>  http://mayavi.sourceforge.net/docs/guide/x967.html#TOOLS

I am using MayaVI for some time, it is reall an excellent tool. The direct call from python looked cumbersome. I'll try to write a script to automate some steps, what could make it more handy.

Another excellent tool for viewing 2-4D array is flounder:
http://www.enel.ucalgary.ca/~vigmond/flounder/

>Best,
>
>Arnd

 Thank you,
  Nadav.



From nadavh at visionsense.com  Thu May  6 06:42:35 2004
From: nadavh at visionsense.com (Nadav Horesh)
Date: Thu, 6 May 2004 12:42:35 +0200
Subject: [SciPy-user] Re: xplt documentation
Message-ID: <07C6A61102C94148B8104D42DE95F7E86DECF0@exchange2k.envision.co.il>




-----Original Message-----
From:	John Hunter [mailto:jdhunter at ace.bsd.uchicago.edu]
Sent:	Tue 04-May-04 15:31
To:	SciPy Users List
Cc:	Nadav Horesh
Subject:	Re: [SciPy-user] Re: xplt documentation
>>>>> "Arnd" == Arnd Baecker <arnd.baecker at web.de> writes:

    Arnd> There is a patch on sourceforge for gnuplot which allows for
    Arnd> bitmap images (ie. matrices). I once tried it out and it is
    Arnd> very nice.  It might take a little time until this gets
    Arnd> integrated as the gnuplot team seems to be a bit busy with
    Arnd> after release issues.

    Arnd> If you really want to go for large matrices have a look at
    Arnd> MayaVi, http://mayavi.sourceforge.net/ In particular
    Arnd> http://mayavi.sourceforge.net/docs/guide/x967.html#TOOLS

>Also the fairly new imshow in matplotlib plots grayscale and
>pseudocolor numeric/numarray arrays.  This replaces the painfully slow
>pcolor which was unusable except for small arrays.  The guts of imshow
>are implemented entirely in numeric and extension could
>handle large arrays efficiently.  If you encounter performance issues,
>I'd like to know about it!

Indeed this is the best opton I have at the moment to fast-create a false-colour maps for large matrices.

>Right now the only false colormap is ColormapJet (see
>examples/pcolor_demo2.py) but you can define your own by subclassing
>matplotlib.colors.Colormap.

I played for some time with the spectrum of colour avaiable in gnuplot's "set palette" command, and finally I picked one that is very close to matplotlib's, so there is nothing really to complain about.

>JDH

  Thanks,
     Nadav.





From travis at enthought.com  Thu May  6 08:27:08 2004
From: travis at enthought.com (Travis N. Vaught)
Date: Thu, 06 May 2004 07:27:08 -0500
Subject: [SciPy-user] ANN: SciPy 2004 Conference - Python for Scientific
	Computing
Message-ID: <409A2F1C.7050801@enthought.com>

Greetings,

The 1st annual *SciPy Conference* will be held this year at Caltech, 
September 2-3, 2004.  As some of you may know, we've experienced great 
participation in two SciPy "Workshops" (with ~70 attendees in both 2002 
and 2003) and this year we're graduating to a "conference."  With the 
prestige of a conference comes the responsibility of a keynote address.  
This year, Jim Hugunin has answered the call and will be speaking to 
kickoff the meeting on Thursday September 2nd.  Jim is the creator of 
Numeric Python, Jython, and co-designer of AspectJ. Jim is currently 
working on IronPython--a fast implementation of Python for .NET and Mono.

Registration is now open.  More information can be found here:

http://www.scipy.org/wikis/scipy04

You may register early online for $100.00.  Registration includes 
breakfast and lunch Thursday & Friday and a very nice dinner Thursday 
night.  After July 16, registration will cost $150.00.

Call for Presenters:
If you are interested in presenting at the conference, you may submit an 
abstract in Plain Text, PDF or MS Word formats to abstracts at scipy.org -- 
the deadline for abstract submission is July 1, 2004.  Papers and/or 
presentation slides are acceptable and are due by August 20, 2004.

We're also planning three days of informal "Coding Sprints" prior to the 
conference -- August 30 to September 1, 2004.  Conference registration 
is not required to participate in the sprints.  Please email the list, 
however, if you plan to attend.  Topics for these sprints will be 
determined via the mailing lists as well, so please submit any 
suggestions for topics to the scipy-user list:

list signup: http://www.scipy.org/mailinglists/
list address: scipy-user at scipy.org

Please forward this announcement to anyone/list that might be interested.

I look forward to seeing you at the conference.

Best Regards,

Travis N. Vaught



From arnd.baecker at web.de  Thu May  6 18:18:34 2004
From: arnd.baecker at web.de (Arnd Baecker)
Date: Fri, 7 May 2004 00:18:34 +0200 (CEST)
Subject: [SciPy-user] help and lazy importer (again or still ?)
In-Reply-To: <Pine.LNX.4.58.0405060143210.20584@scipy.org>
References: <Pine.LNX.4.51.0405051746040.20466@ptpcp2.phy.tu-dresden.de>
	<Pine.LNX.4.58.0405051430410.20584@scipy.org>
	<Pine.LNX.4.51.0405052326100.1305@ptplp09.phy.tu-dresden.de>
	<Pine.LNX.4.58.0405060143210.20584@scipy.org>
Message-ID: <Pine.LNX.4.51.0405062341290.1417@ptplp09.phy.tu-dresden.de>

Hi Pearu,

On Thu, 6 May 2004, Pearu Peterson wrote:

> It turns out I was wrong, the latest patch in CVS makes also
>
>   help("scipy.integrate")
>
> to work.

I just tried it out (>>> scipy.__version__: '0.3.1_280.4183')

>>> help("scipy.integrate")
No module named Image
>>> help("scipy.integrate")
No module named ImageFilter
>>> help("scipy.integrate")
No module named wxPython
>>> help("scipy.integrate")
  [[[ now the expected help shows up ]]]

I.e, only the 4th try leads to the doc-string?

Another one:
>>> help("scipy.integrate.odeint")
No module named Image

>>> help("scipy.integrate.odeint")
No module named odeint

Do you also observe this or did I screw up my installation ?

Best,

Arnd



From oliphant at ee.byu.edu  Thu May  6 21:13:42 2004
From: oliphant at ee.byu.edu (Travis Oliphant)
Date: Thu, 06 May 2004 19:13:42 -0600
Subject: [SciPy-user] Re: Windose and Python
In-Reply-To: <1083889365.6408.37.camel@newton.sandia.gov>
References: <1083889365.6408.37.camel@newton.sandia.gov>
Message-ID: <409AE2C6.1010808@ee.byu.edu>

Todd Pitts from Sandia asked me the following question.

>I have one final question about python on windows.  It seems
>that the non-interactive scripting works well enough.  However, I have
>not found a single interactive interpreter that I could recommend  to
>members of my group without serious reservations.  
>
>I have tried IPython,
>PyCrust (various), IDLE, Using it from within emacs (not cygwin emacs,
>just emacs under windows), PythonWin, etc.  They all have serious
>problems when it comes to usability.  Most don't have tab completion at
>all.   Most, (emacs included) don't work with any plotting package.  I
>have tried gist from scipy and matplotlib (doesn't work with anything
>except straight scripting). 
>
>Is python really this unusable for
>interactive data exploration and modeling under Windows?  
>

I'm forwarding it to these lists so that individuals with more 
experience on Windows than I have can respond to his request. 

What do people use on Windows for interactive work????

-Travis Oliphant






From haase at msg.ucsf.edu  Thu May  6 21:20:28 2004
From: haase at msg.ucsf.edu (Sebastian Haase)
Date: Thu, 6 May 2004 18:20:28 -0700
Subject: [SciPy-user] Re: Windose and Python
In-Reply-To: <409AE2C6.1010808@ee.byu.edu>
References: <1083889365.6408.37.camel@newton.sandia.gov>
	<409AE2C6.1010808@ee.byu.edu>
Message-ID: <200405061820.29022.haase@msg.ucsf.edu>

We (I) like PyCrust quite a lot. What is your complain with this in 
particular ?

- Sebastian Haase

On Thursday 06 May 2004 06:13 pm, Travis Oliphant wrote:
> Todd Pitts from Sandia asked me the following question.
>
> >I have one final question about python on windows.  It seems
> >that the non-interactive scripting works well enough.  However, I have
> >not found a single interactive interpreter that I could recommend  to
> >members of my group without serious reservations.
> >
> >I have tried IPython,
> >PyCrust (various), IDLE, Using it from within emacs (not cygwin emacs,
> >just emacs under windows), PythonWin, etc.  They all have serious
> >problems when it comes to usability.  Most don't have tab completion at
> >all.   Most, (emacs included) don't work with any plotting package.  I
> >have tried gist from scipy and matplotlib (doesn't work with anything
> >except straight scripting).
> >
> >Is python really this unusable for
> >interactive data exploration and modeling under Windows?
>
> I'm forwarding it to these lists so that individuals with more
> experience on Windows than I have can respond to his request.
>
> What do people use on Windows for interactive work????
>
> -Travis Oliphant
>
>
>
>
> _______________________________________________
> SciPy-user mailing list
> SciPy-user at scipy.net
> http://www.scipy.net/mailman/listinfo/scipy-user



From jdhunter at ace.bsd.uchicago.edu  Thu May  6 21:56:25 2004
From: jdhunter at ace.bsd.uchicago.edu (John Hunter)
Date: Thu, 06 May 2004 20:56:25 -0500
Subject: [SciPy-user] Re: Windose and Python
In-Reply-To: <409AE2C6.1010808@ee.byu.edu> (Travis Oliphant's message of
	"Thu, 06 May 2004 19:13:42 -0600")
References: <1083889365.6408.37.camel@newton.sandia.gov>
	<409AE2C6.1010808@ee.byu.edu>
Message-ID: <m2isf9hueu.fsf@mother.paradise.lost>

>>>>> "Travis" == Travis Oliphant <oliphant at ee.byu.edu> writes:

    >> (emacs included) don't work with any plotting package.  I have
    >> tried gist from scipy and matplotlib (doesn't work with
    >> anything except straight scripting). Is python really this
    >> unusable for interactive data exploration and modeling under
    >> Windows?

    Travis> I'm forwarding it to these lists so that individuals with
    Travis> more experience on Windows than I have can respond to his
    Travis> request. What do people use on Windows for interactive
    Travis> work????

Have you tried matplotlib with the TkAgg backend using the standard
python shell, ipython or idle launched with -n?  Most people report
good luck on windows with one of these shells for interactive use,
particularly the first two.  The TkAgg backend is a fairly recent
addition, and a couple of settings in the matplotlib rc file will make
your experience a little more pleasant

backend      : TkAgg     
interactive  : True          
tk.window_focus   : True    # Maintain shell focus for TkAgg

Now when you fire up python or ipython and then import matplotlib,
you'll be in interactive mode using the Tkinter backend.  The window
focus setting is designed to keep your figure from taking the focus
when you issue plotting commands.

Admittedly scripting is the primary way most people use matplotlib,
but we've been working to make the interactive experience better.  So
if it's been a while since you tried it interactively on win32, it may
be worth a second look using a recent release.  It is important to
consult the backends section of the web page to make sure your IDE is
compatible with the backend you are using, however.  Finally, Todd
Miller, who developed the Tk backend, has been very responsive in
fixing known problems, so if you'll let us know what limitations you
find we'll do what we can to fix them up.

John Hunter



From pearu at scipy.org  Thu May  6 23:53:18 2004
From: pearu at scipy.org (Pearu Peterson)
Date: Thu, 6 May 2004 22:53:18 -0500 (CDT)
Subject: [SciPy-user] help and lazy importer (again or still ?)
In-Reply-To: <Pine.LNX.4.51.0405062341290.1417@ptplp09.phy.tu-dresden.de>
References: <Pine.LNX.4.51.0405051746040.20466@ptpcp2.phy.tu-dresden.de>
	<Pine.LNX.4.58.0405051430410.20584@scipy.org>
	<Pine.LNX.4.51.0405052326100.1305@ptplp09.phy.tu-dresden.de>
	<Pine.LNX.4.58.0405060143210.20584@scipy.org>
	<Pine.LNX.4.51.0405062341290.1417@ptplp09.phy.tu-dresden.de>
Message-ID: <Pine.LNX.4.58.0405062247110.20584@scipy.org>



On Fri, 7 May 2004, Arnd Baecker wrote:

> I just tried it out (>>> scipy.__version__: '0.3.1_280.4183')
> 
> >>> help("scipy.integrate")
> No module named Image
> >>> help("scipy.integrate")
> No module named ImageFilter
> >>> help("scipy.integrate")
> No module named wxPython
> >>> help("scipy.integrate")
>   [[[ now the expected help shows up ]]]
> 
> I.e, only the 4th try leads to the doc-string?

Yes, I have noticed that too when wxPython is not installed (Python 2.2).

> Another one:
> >>> help("scipy.integrate.odeint")
> No module named Image
> 
> >>> help("scipy.integrate.odeint")
> No module named odeint

I'll look into it.

Pearu



From fperez at colorado.edu  Fri May  7 00:09:13 2004
From: fperez at colorado.edu (Fernando Perez)
Date: Thu, 06 May 2004 22:09:13 -0600
Subject: [SciPy-user] Re: Windose and Python
In-Reply-To: <409AE2C6.1010808@ee.byu.edu>
References: <1083889365.6408.37.camel@newton.sandia.gov>
	<409AE2C6.1010808@ee.byu.edu>
Message-ID: <409B0BE9.7090305@colorado.edu>

Travis Oliphant wrote:
> Todd Pitts from Sandia asked me the following question.
> 
> 
>>I have one final question about python on windows.  It seems
>>that the non-interactive scripting works well enough.  However, I have
>>not found a single interactive interpreter that I could recommend  to
>>members of my group without serious reservations.  
>>
>>I have tried IPython,
>>PyCrust (various), IDLE, Using it from within emacs (not cygwin emacs,
>>just emacs under windows), PythonWin, etc.  They all have serious
>>problems when it comes to usability.  Most don't have tab completion at
>>all.   Most, (emacs included) don't work with any plotting package.  I
>>have tried gist from scipy and matplotlib (doesn't work with anything
>>except straight scripting). 
>>
>>Is python really this unusable for
>>interactive data exploration and modeling under Windows?  

As the ipython (http://ipython.scipy.org) author I'm obviously biased, but 
Windows users seem fairly happy with it.  Using Gary Bishop's extensions, it 
is possible (it should basically work out of the box, though I don't know 
because I don't use Windows) to get readline and coloring support under a 
normal (non-cygwin) command shell.

Gary's tools are at:
http://sourceforge.net/projects/uncpythontools

I also imagine that using ipython within emacs as your python shell (which 
requires a special python-mode.el and ipython.el, available at 
http://ipython.scipy.org/dist/ipython-emacs-0.3.tgz) must be an option under 
Windows.  I've only used them under Linux, but since this is just regular 
Emacs lisp, I imagine it should be platform-independent.

I hope this helps.

Regards,

Fernando.



From fperez at colorado.edu  Fri May  7 00:09:13 2004
From: fperez at colorado.edu (Fernando Perez)
Date: Thu, 06 May 2004 22:09:13 -0600
Subject: [SciPy-user] Re: Windose and Python
In-Reply-To: <409AE2C6.1010808@ee.byu.edu>
References: <1083889365.6408.37.camel@newton.sandia.gov>
	<409AE2C6.1010808@ee.byu.edu>
Message-ID: <409B0BE9.7090305@colorado.edu>

Travis Oliphant wrote:
> Todd Pitts from Sandia asked me the following question.
> 
> 
>>I have one final question about python on windows.  It seems
>>that the non-interactive scripting works well enough.  However, I have
>>not found a single interactive interpreter that I could recommend  to
>>members of my group without serious reservations.  
>>
>>I have tried IPython,
>>PyCrust (various), IDLE, Using it from within emacs (not cygwin emacs,
>>just emacs under windows), PythonWin, etc.  They all have serious
>>problems when it comes to usability.  Most don't have tab completion at
>>all.   Most, (emacs included) don't work with any plotting package.  I
>>have tried gist from scipy and matplotlib (doesn't work with anything
>>except straight scripting). 
>>
>>Is python really this unusable for
>>interactive data exploration and modeling under Windows?  

As the ipython (http://ipython.scipy.org) author I'm obviously biased, but 
Windows users seem fairly happy with it.  Using Gary Bishop's extensions, it 
is possible (it should basically work out of the box, though I don't know 
because I don't use Windows) to get readline and coloring support under a 
normal (non-cygwin) command shell.

Gary's tools are at:
http://sourceforge.net/projects/uncpythontools

I also imagine that using ipython within emacs as your python shell (which 
requires a special python-mode.el and ipython.el, available at 
http://ipython.scipy.org/dist/ipython-emacs-0.3.tgz) must be an option under 
Windows.  I've only used them under Linux, but since this is just regular 
Emacs lisp, I imagine it should be platform-independent.

I hope this helps.

Regards,

Fernando.



From fperez at colorado.edu  Fri May  7 00:19:57 2004
From: fperez at colorado.edu (Fernando Perez)
Date: Thu, 06 May 2004 22:19:57 -0600
Subject: [SciPy-user] Re: Windose and Python
In-Reply-To: <409AE2C6.1010808@ee.byu.edu>
References: <1083889365.6408.37.camel@newton.sandia.gov>
	<409AE2C6.1010808@ee.byu.edu>
Message-ID: <409B0E6D.5060306@colorado.edu>

Travis Oliphant wrote:
> Todd Pitts from Sandia asked me the following question.

>>all.   Most, (emacs included) don't work with any plotting package.  I
>>have tried gist from scipy and matplotlib (doesn't work with anything
>>except straight scripting). 

Sorry for the 2nd post.  I forgot to mention plotting: ipython includes 
enhanced support for Gnuplot, with modifications to the interactive plotting 
syntax to make it as quick and easy to use as possible.  I use python 100% of 
the time I'm working on scientific code and data exploration, and my 
environment is: Xemacs for heavy editing, a terminal with ipython for 
interactive work, Gnuplot (with ipython's extensions) for 2d plotting and 
Mayavi (http://mayavi.sourceforge.net) for sophisticated data visualization. 
With Gnuplot 4.0's mouse support, it is an extremely convenient tool for fast 
data exploration, capable of publication-quality PostScript output.  Finally, 
for diagram generation and other problems of a graphical but not purely 
'plotting' nature, I have been very happy with PyX.

IPython was designed _specifically_ to make interactive scientific computing 
work as fluid as possible.  It has direct access to the underlying system 
shell, it remembers previous values (like Mathematica's %N variables), and has 
  many other features which you may find useful in this kind of context.

I haven't looked at matplotlib yet (I've been using gnuplot since the days of 
Windows 3.0), but I will very soon, and I have heard excellent things about 
it.  For those already familiar with matlab's syntax, this may be a better 
option than gnuplot.  If there are any problems with ipython's interaction 
with matplotlib, I'll gladly fix them if possible.

Regards,

Fernando.



From fperez at colorado.edu  Fri May  7 00:19:57 2004
From: fperez at colorado.edu (Fernando Perez)
Date: Thu, 06 May 2004 22:19:57 -0600
Subject: [SciPy-user] Re: Windose and Python
In-Reply-To: <409AE2C6.1010808@ee.byu.edu>
References: <1083889365.6408.37.camel@newton.sandia.gov>
	<409AE2C6.1010808@ee.byu.edu>
Message-ID: <409B0E6D.5060306@colorado.edu>

Travis Oliphant wrote:
> Todd Pitts from Sandia asked me the following question.

>>all.   Most, (emacs included) don't work with any plotting package.  I
>>have tried gist from scipy and matplotlib (doesn't work with anything
>>except straight scripting). 

Sorry for the 2nd post.  I forgot to mention plotting: ipython includes 
enhanced support for Gnuplot, with modifications to the interactive plotting 
syntax to make it as quick and easy to use as possible.  I use python 100% of 
the time I'm working on scientific code and data exploration, and my 
environment is: Xemacs for heavy editing, a terminal with ipython for 
interactive work, Gnuplot (with ipython's extensions) for 2d plotting and 
Mayavi (http://mayavi.sourceforge.net) for sophisticated data visualization. 
With Gnuplot 4.0's mouse support, it is an extremely convenient tool for fast 
data exploration, capable of publication-quality PostScript output.  Finally, 
for diagram generation and other problems of a graphical but not purely 
'plotting' nature, I have been very happy with PyX.

IPython was designed _specifically_ to make interactive scientific computing 
work as fluid as possible.  It has direct access to the underlying system 
shell, it remembers previous values (like Mathematica's %N variables), and has 
  many other features which you may find useful in this kind of context.

I haven't looked at matplotlib yet (I've been using gnuplot since the days of 
Windows 3.0), but I will very soon, and I have heard excellent things about 
it.  For those already familiar with matlab's syntax, this may be a better 
option than gnuplot.  If there are any problems with ipython's interaction 
with matplotlib, I'll gladly fix them if possible.

Regards,

Fernando.



From arnd.baecker at web.de  Fri May  7 03:59:03 2004
From: arnd.baecker at web.de (Arnd Baecker)
Date: Fri, 7 May 2004 09:59:03 +0200 (CEST)
Subject: [SciPy-user] scipy on quantian
In-Reply-To: <40991234.6020204@enthought.com>
References: <Pine.LNX.4.51.0405051122020.20466@ptpcp2.phy.tu-dresden.de>
	<Pine.LNX.4.51.0405051713260.20466@ptpcp2.phy.tu-dresden.de>
	<40991234.6020204@enthought.com>
Message-ID: <Pine.LNX.4.51.0405070956210.1417@ptplp09.phy.tu-dresden.de>

Hi,

On Wed, 5 May 2004, eric jones wrote:

>>Also, you can always post a "news" articles from your personal scipy
>>page that will show up on the front page.

I just tried that and have a text in  my
  http://www.scipy.org/Members/baecker/folder_contents
Now, how do I make that appear in the News section on the front page?
(or does this happen automatically after approval ?)

Best,

Arnd



From karthik at james.hut.fi  Fri May  7 04:21:31 2004
From: karthik at james.hut.fi (Karthikesh Raju)
Date: Fri, 7 May 2004 11:21:31 +0300
Subject: [SciPy-user] covariance and other linear algebra functions
Message-ID: <Pine.SGI.4.58.0405071118120.17460@james.hut.fi>

Hi,

Is am searching for a covariance function and other linear algebra
functions like eigen value decomposition. i am interested in both
scipy (havent managed to get scipy installed on the main computation
systems) and also numarray solution.

With warm regards
karthik


-----------------------------------------------------------------------
Karthikesh Raju,		    email: karthik at james.hut.fi
Researcher,			    http://www.cis.hut.fi/karthik
Helsinki University of Technology,  Tel: +358-9-451 5389
Laboratory of Comp. & Info. Sc.,    Fax: +358-9-451 3277
Department of Computer Sc.,
P.O Box 5400, FIN 02015 HUT,
Espoo, FINLAND
-----------------------------------------------------------------------



From eric at enthought.com  Fri May  7 04:37:22 2004
From: eric at enthought.com (eric jones)
Date: Fri, 07 May 2004 03:37:22 -0500
Subject: [SciPy-user] scipy on quantian
In-Reply-To: <Pine.LNX.4.51.0405070956210.1417@ptplp09.phy.tu-dresden.de>
References: <Pine.LNX.4.51.0405051122020.20466@ptpcp2.phy.tu-dresden.de>
	<Pine.LNX.4.51.0405051713260.20466@ptpcp2.phy.tu-dresden.de>
	<40991234.6020204@enthought.com>
	<Pine.LNX.4.51.0405070956210.1417@ptplp09.phy.tu-dresden.de>
Message-ID: <409B4AC2.2080404@enthought.com>

I just "published" it.  I hope this step won't be necessary at some 
point in the future.

eric


Arnd Baecker wrote:

>Hi,
>
>On Wed, 5 May 2004, eric jones wrote:
>
>  
>
>>>Also, you can always post a "news" articles from your personal scipy
>>>page that will show up on the front page.
>>>      
>>>
>
>I just tried that and have a text in  my
>  http://www.scipy.org/Members/baecker/folder_contents
>Now, how do I make that appear in the News section on the front page?
>(or does this happen automatically after approval ?)
>
>Best,
>
>Arnd
>
>_______________________________________________
>SciPy-user mailing list
>SciPy-user at scipy.net
>http://www.scipy.net/mailman/listinfo/scipy-user
>  
>




From funkmeister at lynxseismicdata.com  Fri May  7 18:47:11 2004
From: funkmeister at lynxseismicdata.com (Funkmeister)
Date: Fri, 7 May 2004 16:47:11 -0600
Subject: [SciPy-user] Logistic Regression Modules?
Message-ID: <7A48F014-A078-11D8-9A2A-000A95B074E0@lynxseismicdata.com>

Hello SciPy Users, I am an experienced python user, but I am rather new 
to SciPy and a little rusty with Stats applications so I apologize now 
for asking a possibly obvious question.

I am working with the GRASS GIS system and need to do some logistic 
regression computations. Are there any modules in SciPy to do this, or 
do I need to write them? In the documentation i found genlogistic and 
logistic under the statistical functions. Would I be able to use these 
functions to do the regression analysis?

Thanks In Advance,
Craig



From rossini at blindglobe.net  Fri May  7 20:54:24 2004
From: rossini at blindglobe.net (A.J. Rossini)
Date: Fri, 07 May 2004 17:54:24 -0700
Subject: [SciPy-user] Logistic Regression Modules?
In-Reply-To: <7A48F014-A078-11D8-9A2A-000A95B074E0@lynxseismicdata.com>
	(funkmeister@lynxseismicdata.com's
	message of "Fri, 7 May 2004 16:47:11 -0600")
References: <7A48F014-A078-11D8-9A2A-000A95B074E0@lynxseismicdata.com>
Message-ID: <853c6beo1r.fsf@servant.blindglobe.net>

Funkmeister <funkmeister at lynxseismicdata.com> writes:

> Hello SciPy Users, I am an experienced python user, but I am rather
> new to SciPy and a little rusty with Stats applications so I apologize
> now for asking a possibly obvious question.
>
> I am working with the GRASS GIS system and need to do some logistic
> regression computations. Are there any modules in SciPy to do this, or
> do I need to write them? In the documentation i found genlogistic and
> logistic under the statistical functions. Would I be able to use these
> functions to do the regression analysis?

I'm guessing not.  The simplest approach would be to wrap a call to
R's glm function via Rpy.

The second simplest would be to plug in the likelihood function and
optimize (which is usually how one used to do this -- I wrote one for
my thesis, via newton-raphson, eons ago (well, mid 90's). 

You could also simulate the optimization via EM, but that's a pretty
slow approach.

best,
-tony

-- 
rossini at u.washington.edu            http://www.analytics.washington.edu/ 
Biomedical and Health Informatics   University of Washington
Biostatistics, SCHARP/HVTN          Fred Hutchinson Cancer Research Center
UW (Tu/Th/F): 206-616-7630 FAX=206-543-3461 | Voicemail is unreliable
FHCRC  (M/W): 206-667-7025 FAX=206-667-4812 | use Email

CONFIDENTIALITY NOTICE: This e-mail message and any attachments may be
confidential and privileged. If you received this message in error,
please destroy it and notify the sender. Thank you.



From scipy at zunzun.com  Fri May  7 22:27:27 2004
From: scipy at zunzun.com (James R. Phillips)
Date: Fri, 7 May 2004 22:27:27 -400
Subject: [SciPy-user] Enthought 0.3 mingw32 weave in WinXP now working
Message-ID: <409c458f6addc0.17626783@mercury.sabren.com>

Whew.  Here are the magic chants from python.org:

http://docs.python.org/inst/tweak-flags.html#SECTION000622000000000000000

That one was painful to figure out.

     James Phillips
     http://zunzun.com



From eric at enthought.com  Fri May  7 22:46:47 2004
From: eric at enthought.com (eric jones)
Date: Fri, 07 May 2004 21:46:47 -0500
Subject: [SciPy-user] Enthought 0.3 mingw32 weave in WinXP now working
In-Reply-To: <409c458f6addc0.17626783@mercury.sabren.com>
References: <409c458f6addc0.17626783@mercury.sabren.com>
Message-ID: <409C4A17.20704@enthought.com>

This should happen automatically when using weave.  It has the creation 
of the libpython.a step built in if it doesn't find it.

Here is what I see (after deleting my existing libpython23.a file in 
c:\python23\libs):

c:\wrk\proava_new\src>python
Enthought Edition build 1057
Python 2.3.3 (#51, Feb 16 2004, 04:07:52) [MSC v.1200 32 bit (Intel)] on 
win32
Type "help", "copyright", "credits" or "license" for more information.
 >>> import weave
 >>> weave.inline(" ",compiler='gcc')
file changed
mingw32
Building import library: "C:\Python23\libs\libpython23.a"

Not sure why it doesn't fly on your machine.

eric



James R. Phillips wrote:

>Whew.  Here are the magic chants from python.org:
>
>http://docs.python.org/inst/tweak-flags.html#SECTION000622000000000000000
>
>That one was painful to figure out.
>
>     James Phillips
>     http://zunzun.com
>
>_______________________________________________
>SciPy-user mailing list
>SciPy-user at scipy.net
>http://www.scipy.net/mailman/listinfo/scipy-user
>  
>




From eric at enthought.com  Sat May  8 11:41:31 2004
From: eric at enthought.com (eric jones)
Date: Sat, 08 May 2004 10:41:31 -0500
Subject: [Fwd: Re: [SciPy-user] Enthought 0.3 mingw32 weave in WinXP now
	working]
Message-ID: <409CFFAB.1050907@enthought.com>

Oops. responded only to James...
-------------- next part --------------
An embedded message was scrubbed...
From: eric jones <eric at enthought.com>
Subject: Re: [SciPy-user] Enthought 0.3 mingw32 weave in WinXP now working
Date: Sat, 08 May 2004 10:27:51 -0500
Size: 2032
URL: <http://mail.scipy.org/pipermail/scipy-user/attachments/20040508/5b7fdc45/attachment.mht>

From pearu at scipy.org  Sun May  9 03:23:17 2004
From: pearu at scipy.org (Pearu Peterson)
Date: Sun, 9 May 2004 02:23:17 -0500 (CDT)
Subject: [SciPy-user] help and lazy importer (again or still ?)
In-Reply-To: <Pine.LNX.4.58.0405062247110.20584@scipy.org>
References: <Pine.LNX.4.51.0405051746040.20466@ptpcp2.phy.tu-dresden.de>
	<Pine.LNX.4.58.0405051430410.20584@scipy.org>
	<Pine.LNX.4.51.0405052326100.1305@ptplp09.phy.tu-dresden.de>
	<Pine.LNX.4.58.0405060143210.20584@scipy.org>
	<Pine.LNX.4.51.0405062341290.1417@ptplp09.phy.tu-dresden.de>
	<Pine.LNX.4.58.0405062247110.20584@scipy.org>
Message-ID: <Pine.LNX.4.58.0405090220400.20584@scipy.org>



On Thu, 6 May 2004, Pearu Peterson wrote:

> On Fri, 7 May 2004, Arnd Baecker wrote:
> 
> > I just tried it out (>>> scipy.__version__: '0.3.1_280.4183')
> > 
> > >>> help("scipy.integrate")
> > No module named Image
> > >>> help("scipy.integrate")
> > No module named ImageFilter
> > >>> help("scipy.integrate")
> > No module named wxPython
> > >>> help("scipy.integrate")
> >   [[[ now the expected help shows up ]]]
> > 
> > I.e, only the 4th try leads to the doc-string?
> 
> Yes, I have noticed that too when wxPython is not installed (Python 2.2).
> 
> > Another one:
> > >>> help("scipy.integrate.odeint")
> > No module named Image
> > 
> > >>> help("scipy.integrate.odeint")
> > No module named odeint

The above issues are now fixed in CVS. When PIL or wxPython are not 
installed then expect the following messages when importing scipy:

/usr/local/lib/python2.2/site-packages/scipy/pilutil.pyc No module named 
PIL
/usr/local/lib/python2.2/site-packages/scipy/plt/interface.pyc No module 
named wxPython
/usr/local/lib/python2.2/site-packages/scipy/plt/__init__.pyc No module 
named wxPython


Pearu



From gazzar at email.com  Mon May 10 00:12:58 2004
From: gazzar at email.com (Gary Ruben)
Date: Mon, 10 May 2004 14:12:58 +1000
Subject: [SciPy-user] Announce: ErrorVal.py 1.0 for error-bounds calculations
Message-ID: <20040510041258.DA8523CE181@ws3-4.us4.outblaze.com>

I finally got around to putting my errorbar calculation module on my website.
<http://users.bigpond.net.au/gazzar/python.html>
This should be of interest to Numeric Python users who deal with calculation of error-bounds on experimental data.

Synopsis:

A module providing a Python number type class and helper functions to ease
the task of computing error bounds on experimental data values.

This module defines an abstract data type, Err, which carries a central/prime
value and upper and lower error bounds.
Helper functions are provided to allow Python Numeric rank-1 arrays of Err objects
to be constructed with ease and to apply Numeric Ufuncs.
Written under Python 2.3.3 and Numeric 23.0

Example of usage:

upperPressure = ArrayOfErr([909., 802., 677., 585., 560., 548.], 1.0)
lowerPressure = ArrayOfErr([144., 246., 378., 469., 493., 505.], 1.0)
pressureDiff = upperPressure - lowerPressure
V_RStandard = ArrayOfErr([2.016, 2.016, 2.020, 2.017, 2.021, 2.019], 0.001)
R = 100.2                                   # standard resistor value [Ohm]
I = V_RStandard / R                         # current [A]
V_RAB = ArrayOfErr([6.167, 6.168, 6.170, 6.160, (6.153, 0.02), (5.894, 0.01)], 0.002)
R_AB = V_RAB / I
logR_AB = ApplyUfuncToErr(R_AB, log10)      # This means log10(R_AB)
print PrimeVals(logR_AB)
print MinVals(logR_AB)
print MaxVals(logR_AB)


Enjoy,
Gary Ruben

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



From elcortostrash at gmx.net  Mon May 10 17:37:11 2004
From: elcortostrash at gmx.net (el corto)
Date: Mon, 10 May 2004 23:37:11 +0200 (MEST)
Subject: [SciPy-user] Re: Windose and Python
References: <409AE2C6.1010808@ee.byu.edu>
Message-ID: <26281.1084225031@www48.gmx.net>

> Todd Pitts from Sandia asked me the following question.
> 
> >I have one final question about python on windows.  It seems
> >that the non-interactive scripting works well enough.  However, I have
> >not found a single interactive interpreter that I could recommend  to
> >members of my group without serious reservations.  
> >
> >I have tried IPython,
> >PyCrust (various), IDLE, Using it from within emacs (not cygwin emacs,
> >just emacs under windows), PythonWin, etc.  They all have serious
> >problems when it comes to usability.  
> >Most don't have tab completion at
> >all.   

This is at least not true for PythonWin. E.g. when you type

from scipy.

a popup shows up when you hit the "." offering several completion options.

I've tested PyCrust and IDLE and found PythonWin the best one for me. 

> >Most, (emacs included) don't work with any plotting package.  I
> >have tried gist from scipy and matplotlib (doesn't work with anything
> >except straight scripting).

I use scipy.xplt with absolutely no probs.
 
> >
> >Is python really this unusable for
> >interactive data exploration and modeling under Windows? 

No :)

> >
> 
> I'm forwarding it to these lists so that individuals with more 
> experience on Windows than I have can respond to his request. 
> 
> What do people use on Windows for interactive work????
> 
> -Travis Oliphant
> 
> 
> 
> 
> _______________________________________________
> SciPy-user mailing list
> SciPy-user at scipy.net
> http://www.scipy.net/mailman/listinfo/scipy-user
> 

-- 
"In science one tries to tell people, in such a way as to be understood by
everyone, something that no one ever knew before. But in poetry, it's the
exact opposite." -Paul Dirac (1902-1984)

"Sie haben neue Mails!" - Die GMX Toolbar informiert Sie beim Surfen!
Jetzt aktivieren unter http://www.gmx.net/info



From arnd.baecker at web.de  Tue May 11 03:00:12 2004
From: arnd.baecker at web.de (Arnd Baecker)
Date: Tue, 11 May 2004 09:00:12 +0200 (CEST)
Subject: [SciPy-user] help and lazy importer (again or still ?)
In-Reply-To: <Pine.LNX.4.58.0405090220400.20584@scipy.org>
References: <Pine.LNX.4.51.0405051746040.20466@ptpcp2.phy.tu-dresden.de>
	<Pine.LNX.4.58.0405051430410.20584@scipy.org>
	<Pine.LNX.4.51.0405052326100.1305@ptplp09.phy.tu-dresden.de>
	<Pine.LNX.4.58.0405060143210.20584@scipy.org>
	<Pine.LNX.4.51.0405062341290.1417@ptplp09.phy.tu-dresden.de>
	<Pine.LNX.4.58.0405090220400.20584@scipy.org>
Message-ID: <Pine.LNX.4.51.0405110803420.1235@ptplp09.phy.tu-dresden.de>

Dear Pearu,

I checked out the CVS version and there seem to be
some further issues (each starting with a fresh python
session):

Case 1 (this works now - thanks!!!)
-----------------------------------

help("scipy.integrate.odeint")
from scipy.integrate import odeint
help("scipy.integrate.odeint")


Case 2
------

help("scipy.integrateMISPELLED.odeint")
help("scipy.integrate.odeint")

gives

Python 2.3.3 (#1, May  3 2004, 16:38:25)
[GCC 3.3.3 (Debian)] on linux2
Type "help", "copyright", "credits" or "license" for more information.
>>> help("scipy.integrateMISPELLED.odeint")
/home/python/PYTHON_New/lib/python2.3/site-packages/scipy/pilutil.pyc No
module named PIL
no Python documentation found for 'scipy.integrateMISPELLED.odeint'

>>> help("scipy.integrate.odeint")
ppimport('scipy.integrate.odeint') caller locals:
__name__=__main__
No module named odeint

and no further repetition of the call helps here ;-).
Doing then
  help("scipy.xplt")
works, and after this
  help("scipy.integrate.odeint")
works as well.


Case 3
------

from scipy.integrate import odeint

def f(x):
   """doc string for f(x)"""
   return x*x

>>> help(f)
[[gives the help on f and then, after returning from the pager: ]]
/home/python/PYTHON_New/lib/python2.3/site-packages/scipy/plt/interface.pyc
No module named wxPython
/home/python/PYTHON_New/lib/python2.3/site-packages/scipy/plt/__init__.pyc
No module named wxPython

>From a user perspective I would not have thought to see this here ;-)


Sorry for being a pain here, but maybe
you (again) have a solution for the above points.

Many thanks, Arnd



From karthik at james.hut.fi  Tue May 11 05:26:21 2004
From: karthik at james.hut.fi (Karthikesh Raju)
Date: Tue, 11 May 2004 12:26:21 +0300
Subject: [SciPy-user] simple question about digonal matrices
Message-ID: <Pine.SGI.4.58.0405111223000.96956@james.hut.fi>

Hi All,

How does one do something as mundane as assigning value to a diagonal
vector.

D = array([1,2,3])

D_with_zeros = array([[0,0,0],
                     [0,0,0],
                     [0,0,0]])

now i want D in the diagonal of D_with_zeros, i was trying something like:
numarray.diagonal(D_with_zeros) = D

numarray is not happy with my assignment?

Warm regards
karthik








-----------------------------------------------------------------------
Karthikesh Raju,		    email: karthik at james.hut.fi
Researcher,			    http://www.cis.hut.fi/karthik
Helsinki University of Technology,  Tel: +358-9-451 5389
Laboratory of Comp. & Info. Sc.,    Fax: +358-9-451 3277
Department of Computer Sc.,
P.O Box 5400, FIN 02015 HUT,
Espoo, FINLAND
-----------------------------------------------------------------------



From karthik at james.hut.fi  Tue May 11 05:27:41 2004
From: karthik at james.hut.fi (Karthikesh Raju)
Date: Tue, 11 May 2004 12:27:41 +0300
Subject: [SciPy-user] simple question about digonal matrices
Message-ID: <Pine.SGI.4.58.0405111223000.96956@james.hut.fi>

Hi All,

How does one do something as mundane as assigning value to a diagonal
vector.

D = array([1,2,3])

D_with_zeros = array([[0,0,0],
                     [0,0,0],
                     [0,0,0]])

now i want D in the diagonal of D_with_zeros, i was trying something like:
numarray.diagonal(D_with_zeros) = D

numarray is not happy with my assignment?

Warm regards
karthik








-----------------------------------------------------------------------
Karthikesh Raju,		    email: karthik at james.hut.fi
Researcher,			    http://www.cis.hut.fi/karthik
Helsinki University of Technology,  Tel: +358-9-451 5389
Laboratory of Comp. & Info. Sc.,    Fax: +358-9-451 3277
Department of Computer Sc.,
P.O Box 5400, FIN 02015 HUT,
Espoo, FINLAND
-----------------------------------------------------------------------



From pearu at scipy.org  Tue May 11 09:05:03 2004
From: pearu at scipy.org (Pearu Peterson)
Date: Tue, 11 May 2004 08:05:03 -0500 (CDT)
Subject: [SciPy-user] help and lazy importer (again or still ?)
In-Reply-To: <Pine.LNX.4.51.0405110803420.1235@ptplp09.phy.tu-dresden.de>
References: <Pine.LNX.4.51.0405051746040.20466@ptpcp2.phy.tu-dresden.de>
	<Pine.LNX.4.58.0405051430410.20584@scipy.org>
	<Pine.LNX.4.51.0405052326100.1305@ptplp09.phy.tu-dresden.de>
	<Pine.LNX.4.58.0405060143210.20584@scipy.org>
	<Pine.LNX.4.51.0405062341290.1417@ptplp09.phy.tu-dresden.de>
	<Pine.LNX.4.58.0405090220400.20584@scipy.org>
	<Pine.LNX.4.51.0405110803420.1235@ptplp09.phy.tu-dresden.de>
Message-ID: <Pine.LNX.4.58.0405110745080.12352@scipy.org>


Hi Arnd,

On Tue, 11 May 2004, Arnd Baecker wrote:

> I checked out the CVS version and there seem to be
> some further issues (each starting with a fresh python
> session):
> 
> Case 2
> ------
> 
> help("scipy.integrateMISPELLED.odeint")
> help("scipy.integrate.odeint")
> 
> gives
> 
> Python 2.3.3 (#1, May  3 2004, 16:38:25)
> [GCC 3.3.3 (Debian)] on linux2
> Type "help", "copyright", "credits" or "license" for more information.
> >>> help("scipy.integrateMISPELLED.odeint")
> /home/python/PYTHON_New/lib/python2.3/site-packages/scipy/pilutil.pyc No
> module named PIL
> no Python documentation found for 'scipy.integrateMISPELLED.odeint'
> 
> >>> help("scipy.integrate.odeint")
> ppimport('scipy.integrate.odeint') caller locals:
> __name__=__main__
> No module named odeint
> 
> and no further repetition of the call helps here ;-).

Fixed in CVS.

> Case 3
> ------
> 
> from scipy.integrate import odeint
> 
> def f(x):
>    """doc string for f(x)"""
>    return x*x
> 
> >>> help(f)
> [[gives the help on f and then, after returning from the pager: ]]
> /home/python/PYTHON_New/lib/python2.3/site-packages/scipy/plt/interface.pyc
> No module named wxPython
> /home/python/PYTHON_New/lib/python2.3/site-packages/scipy/plt/__init__.pyc
> No module named wxPython
> 
> >From a user perspective I would not have thought to see this here ;-)

I wouldn't disable these messages yet as they are informative for users 
who try to use scipy.plt module (that will not work without wxPython) as 
well as they appear only once. Maybe these messages should appear as 
warnings, though, then one can disable them using python warnings system..

> Sorry for being a pain here, but maybe you (again) have a solution for 
> the above points.

No problem at all. Your feedback is very much appreciated. Keep sending 
bug reports also in future;-)

There is also Scipy bugtracker, http://www.scipy.org/bugtracker, that 
might be the place to save bugreports. Unfortunately, from the developers 
point of view, it is pain to use the current Roundup: almost all issues 
that I have fixed there and set 'resolved', have appeared as 'chatting' 
again after few days, either because of some bug in Roundup itself or 
because of some 'anonymous' user who cannot stand 'resolved' issues:(
Hopefully 'anonymous' users will be disabled in Roundup someday..

Thanks,
Pearu



From gazzar at email.com  Tue May 11 09:25:19 2004
From: gazzar at email.com (Gary Ruben)
Date: Tue, 11 May 2004 23:25:19 +1000
Subject: [SciPy-user] simple question about digonal matrices
Message-ID: <20040511132519.3F1EC1CE303@ws3-6.us4.outblaze.com>

Hi Karthik,
FYI, there's a numpy discussion list here <http://lists.sourceforge.net/lists/listinfo/numpy-discussion> which is probably a more appropriate place to post a numpy-specific question.

As for your problem, here's one way to do it:

>>> from numarray import *
>>> D = array([1,2,3])
>>> D_with_zeros = array([[0,0,0],
		      [0,0,0],
		      [0,0,0]])
>>> where(identity(3),D,D_with_zeros)
array([[1, 0, 0],
       [0, 2, 0],
       [0, 0, 3]])


HTH,
Gary Ruben

----- Original Message -----
From: Karthikesh Raju <karthik at james.hut.fi>
Date: Tue, 11 May 2004 12:26:21 +0300
To: scipy-user at scipy.net
Subject: [SciPy-user] simple question about digonal matrices

> Hi All,
> 
> How does one do something as mundane as assigning value to a diagonal
> vector.
> 
> D = array([1,2,3])
> 
> D_with_zeros = array([[0,0,0],
>                      [0,0,0],
>                      [0,0,0]])
> 
> now i want D in the diagonal of D_with_zeros, i was trying something like:
> numarray.diagonal(D_with_zeros) = D
> 
> numarray is not happy with my assignment?
> 
> Warm regards
> karthik

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



From arnd.baecker at web.de  Tue May 11 09:27:42 2004
From: arnd.baecker at web.de (Arnd Baecker)
Date: Tue, 11 May 2004 15:27:42 +0200 (CEST)
Subject: [SciPy-user] help and lazy importer (again or still ?)
In-Reply-To: <Pine.LNX.4.58.0405110745080.12352@scipy.org>
References: <Pine.LNX.4.51.0405051746040.20466@ptpcp2.phy.tu-dresden.de>
	<Pine.LNX.4.58.0405051430410.20584@scipy.org>
	<Pine.LNX.4.51.0405052326100.1305@ptplp09.phy.tu-dresden.de>
	<Pine.LNX.4.58.0405060143210.20584@scipy.org>
	<Pine.LNX.4.51.0405062341290.1417@ptplp09.phy.tu-dresden.de>
	<Pine.LNX.4.58.0405090220400.20584@scipy.org>
	<Pine.LNX.4.51.0405110803420.1235@ptplp09.phy.tu-dresden.de>
	<Pine.LNX.4.58.0405110745080.12352@scipy.org>
Message-ID: <Pine.LNX.4.51.0405111514370.31277@ptpcp2.phy.tu-dresden.de>

Hi Pearu,

On Tue, 11 May 2004, Pearu Peterson wrote:
[...]
> > Case 3
> > ------
> >
> > from scipy.integrate import odeint
> >
> > def f(x):
> >    """doc string for f(x)"""
> >    return x*x
> >
> > >>> help(f)
> > [[gives the help on f and then, after returning from the pager: ]]
> > /home/python/PYTHON_New/lib/python2.3/site-packages/scipy/plt/interface.pyc
> > No module named wxPython
> > /home/python/PYTHON_New/lib/python2.3/site-packages/scipy/plt/__init__.pyc
> > No module named wxPython
> >
> > >From a user perspective I would not have thought to see this here ;-)
>
> I wouldn't disable these messages yet as they are informative for users
> who try to use scipy.plt module (that will not work without wxPython) as
> well as they appear only once. Maybe these messages should appear as
> warnings, though, then one can disable them using python warnings system..

I am not fully convinced ;-): If a user asks for help on a function
which does not contain anything related to scipy or wxPython
he will will be confused about the message and will
have a hard time to find out where the complaint comes from.
Somehow the invocation of "help" on f triggers to import
the scipy stuff and that's where the warning comes from
(Hope this is the right explanation).

In addition, there is (from the users perspective) no place where
scipy.plt is called. Wouldn't it be better to defer any complaints
about a missing wxPython (and PIL in other situations)
when scipy.plt is actually used?
(Ensuring the availability of wxPython/PIL is more a thing
of the installation than that it should pop on every import of scipy,
I think.)
((Honestly, normally I would not see these messages anyway
as I have PIL and wxPython installed, but not for these tests;-))

> > Sorry for being a pain here, but maybe you (again) have a solution for
> > the above points.
>
> No problem at all. Your feedback is very much appreciated. Keep sending
> bug reports also in future;-)

I will ;-)

> There is also Scipy bugtracker, http://www.scipy.org/bugtracker, that
> might be the place to save bugreports. Unfortunately, from the developers
> point of view, it is pain to use the current Roundup: almost all issues
> that I have fixed there and set 'resolved', have appeared as 'chatting'
> again after few days, either because of some bug in Roundup itself or
> because of some 'anonymous' user who cannot stand 'resolved' issues:(

Some people seem to have just too much time.

> Hopefully 'anonymous' users will be disabled in Roundup someday..

Whenever this works properly just let me know, and I will post there
if you prefer that.

Many thanks,

Arnd



From perry at stsci.edu  Tue May 11 11:09:50 2004
From: perry at stsci.edu (Perry Greenfield)
Date: Tue, 11 May 2004 11:09:50 -0400
Subject: [SciPy-user] simple question about digonal matrices
In-Reply-To: <Pine.SGI.4.58.0405111223000.96956@james.hut.fi>
Message-ID: <CEELJPECNGEGKFNDLHFFEEGPCDAA.perry@stsci.edu>

Karthikesh Raju wrote:

> Hi All,
>
> How does one do something as mundane as assigning value to a diagonal
> vector.
>
> D = array([1,2,3])
>
> D_with_zeros = array([[0,0,0],
>                      [0,0,0],
>                      [0,0,0]])
>
> now i want D in the diagonal of D_with_zeros, i was trying something like:
> numarray.diagonal(D_with_zeros) = D
>
> numarray is not happy with my assignment?
>
I don't believe that this is a numarray problem but python syntax
issue. You can't assign to a function call (that is the error
message you got, right?). It seems that you are asking for
a way of creating an array that is a view of the diagonal
of a matrix such that if you assign to elements of that diagonal
vector, the matrix will also be updated. With either Numeric
or numarray, it is possible to create such an array so you can
do that but it involves a couple steps (one could write a function
to do that for you in one step). It would have to work something
like this:

diagonalview(D_with_zeros)[:] = D

Note that you must use a slice assignment for this to work.
Note that this won't work for the diagonal function since
it returns a copy of the diagonal elements and so changing
those doesn't change the matrix.

You could define diagonalview as:

def diagonalview(arr)
	return arr.flat[::arr.shape[1]+1]

(Note that this won't work for noncontinguous arrays and needs
some error checking)

Another approach usable with numarray is:

ind = arange(D_with_zeros.shape[1])
D_with_zeros[ind, ind] = D



From pearu at scipy.org  Tue May 11 15:56:26 2004
From: pearu at scipy.org (Pearu Peterson)
Date: Tue, 11 May 2004 14:56:26 -0500 (CDT)
Subject: [SciPy-user] help and lazy importer (again or still ?)
In-Reply-To: <Pine.LNX.4.51.0405111514370.31277@ptpcp2.phy.tu-dresden.de>
References: <Pine.LNX.4.51.0405051746040.20466@ptpcp2.phy.tu-dresden.de>
	<Pine.LNX.4.58.0405051430410.20584@scipy.org>
	<Pine.LNX.4.51.0405052326100.1305@ptplp09.phy.tu-dresden.de>
	<Pine.LNX.4.58.0405060143210.20584@scipy.org>
	<Pine.LNX.4.51.0405062341290.1417@ptplp09.phy.tu-dresden.de>
	<Pine.LNX.4.58.0405090220400.20584@scipy.org>
	<Pine.LNX.4.51.0405110803420.1235@ptplp09.phy.tu-dresden.de>
	<Pine.LNX.4.58.0405110745080.12352@scipy.org>
	<Pine.LNX.4.51.0405111514370.31277@ptpcp2.phy.tu-dresden.de>
Message-ID: <Pine.LNX.4.58.0405111443350.12352@scipy.org>



On Tue, 11 May 2004, Arnd Baecker wrote:

> On Tue, 11 May 2004, Pearu Peterson wrote:
> [...]
> > > Case 3
> > > ------
> > >
> > > from scipy.integrate import odeint
> > >
> > > def f(x):
> > >    """doc string for f(x)"""
> > >    return x*x
> > >
> > > >>> help(f)
> > > [[gives the help on f and then, after returning from the pager: ]]
> > > /home/python/PYTHON_New/lib/python2.3/site-packages/scipy/plt/interface.pyc
> > > No module named wxPython
> > > /home/python/PYTHON_New/lib/python2.3/site-packages/scipy/plt/__init__.pyc
> > > No module named wxPython
> > >
> > > >From a user perspective I would not have thought to see this here ;-)
> >
> > I wouldn't disable these messages yet as they are informative for users
> > who try to use scipy.plt module (that will not work without wxPython) as
> > well as they appear only once. Maybe these messages should appear as
> > warnings, though, then one can disable them using python warnings system..
> 
> I am not fully convinced ;-):
> ....

Ok, ok ;) The issue is fixed in CVS. Well, you won't see 'No module named 
PIL' messages anymore until trying to use PIL stuff. Sometimes you might 
see 'No module named wxPython', (e.g. when asking help('scipy')) but not 
with help(f) or help('scipy.linalg'). I'll fix this  sometimes later 
(feel free to file a bug report to roundup so that it won't be 
forgotten).

Thanks,
Pearu



From jl at dmi.dk  Wed May 12 10:14:27 2004
From: jl at dmi.dk (Jesper Larsen)
Date: Wed, 12 May 2004 16:14:27 +0200
Subject: [SciPy-user] High-pass filtering
Message-ID: <200405121614.27707.jl@dmi.dk>

Dear scipy mailing list,

I would like to perform a high-pass filtering of a simple timeseries of 
floats. I'm not really sure how to do that using numarray and scipy. Can 
anyone post a simple example of how to do that? I have no special preferences 
for window type.

Kind regards,
Jesper



From oliphant at ee.byu.edu  Wed May 12 16:56:10 2004
From: oliphant at ee.byu.edu (Travis Oliphant)
Date: Wed, 12 May 2004 14:56:10 -0600
Subject: [SciPy-user] High-pass filtering
In-Reply-To: <200405121614.27707.jl@dmi.dk>
References: <200405121614.27707.jl@dmi.dk>
Message-ID: <40A28F6A.6020901@ee.byu.edu>

Jesper Larsen wrote:

>Dear scipy mailing list,
>
>I would like to perform a high-pass filtering of a simple timeseries of 
>floats. I'm not really sure how to do that using numarray and scipy. Can 
>anyone post a simple example of how to do that? I have no special preferences 
>for window type.
>
>  
>

SciPy uses Numeric and doesn't require numarray (it can use numarray 
objects as inputs which will be treated as Python sequences and can be 
slow).

What kind of high-pass filter are you looking for.  Simple FIR filters 
are quite simple to design and give linear phase.  Other IIR filters are 
available.  I'll show you how to design a high-pass FIR filter using the 
remez algorithm (you could also look at signal.firwin for windowed 
filter design (but remez is usually a better solution).

Let T be the sample distance and must be defined

F = 1.0/T  
bands = array([0,0.5,0.6,1])*F/2  # this places band-edges
gain = [0,1]  # high-pass filter
N = 25  # length of FIR filter
b = signal.remez(N, bands, gain,Hz=F)

w,h = signal.freqz(b,1)  # this will show you the filter
xplt.plot(w/pi*F/2,abs(h))

To filter a time series you can use

out = signal.lfilter(b,[1],in_)

Or simply
out = signal.convolve(in_,b)[:-(N-1)]

You can get different transients using

out = signal.convolve(in_,b,'same') (cuts off on both ends)


Good luck,

-Travis O.




From scottbray83 at hotmail.com  Wed May 12 22:17:24 2004
From: scottbray83 at hotmail.com (Scott Bray)
Date: Thu, 13 May 2004 12:17:24 +1000
Subject: [SciPy-user] normalizing data/distributions
Message-ID: <BAY12-F29e3nJIxA0hk00012c7d@hotmail.com>

Hey everyone,

i am working on a statistics type project for university study. i have a set 
of data, have built a discrete probability distribution from this data 
(using the cauchy distribution) and now want to normalize it. currently, the 
area of the distribution is not equal to one. i have been trying to find 
literature about how to normalize, but have been unsuccessful (and what i 
have found, i am unsure on the validity). some say to normalize the data 
points by:

(data point - sample mean) / sample std

others say to multiply by a normalising constant that is "chosen" to make 
the area equal to one. i tried this by just scaling the area to equal one.

i'm sorry if this is unclear, but if you have done this sort of thing 
before, i would REALLY appreciate some help. just ask me some questions if 
needs be.

Thanks
Scott

_________________________________________________________________
What's your house worth? Click here to find out:  
http://www.ninemsn.realestate.com.au



From rkern at ucsd.edu  Wed May 12 22:57:57 2004
From: rkern at ucsd.edu (Robert Kern)
Date: Wed, 12 May 2004 19:57:57 -0700
Subject: [SciPy-user] normalizing data/distributions
In-Reply-To: <BAY12-F29e3nJIxA0hk00012c7d@hotmail.com>
References: <BAY12-F29e3nJIxA0hk00012c7d@hotmail.com>
Message-ID: <40A2E435.50503@ucsd.edu>

Scott Bray wrote:

> Hey everyone,
> 
> i am working on a statistics type project for university study. i have a 
> set of data, have built a discrete probability distribution from this 
> data (using the cauchy distribution) and now want to normalize it.

I'm sorry, but this doesn't make sense. How does one build a discrete 
probability distribution using the Cauchy distribution and the data? 
Cauchy is a continuous distribution. Do you mean parameterized? 
Histogrammed?

> currently, the area of the distribution is not equal to one. i have been 
> trying to find literature about how to normalize, but have been 
> unsuccessful (and what i have found, i am unsure on the validity). some 
> say to normalize the data points by:
> 
> (data point - sample mean) / sample std

The meaning of the word "normalization" varies with context. If you had 
reason to believe the data were Gaussian, this transformation reduces 
the data to a standard ("normal") form. Not what you want, I think.

> others say to multiply by a normalising constant that is "chosen" to 
> make the area equal to one. i tried this by just scaling the area to 
> equal one.

Sounds about right. I'm not sure what you're talking about, though (area 
of what exactly? how are you calculating this thing?).

> i'm sorry if this is unclear, but if you have done this sort of thing 
> before, i would REALLY appreciate some help. just ask me some questions 
> if needs be.

Sure thing, but let's take the statistics issues off-list into private 
email until we get to SciPy issues.

> Thanks
> Scott

-- 
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 rossini at blindglobe.net  Wed May 12 23:14:07 2004
From: rossini at blindglobe.net (A.J. Rossini)
Date: Wed, 12 May 2004 20:14:07 -0700
Subject: [SciPy-user] normalizing data/distributions
In-Reply-To: <BAY12-F29e3nJIxA0hk00012c7d@hotmail.com> (Scott Bray's message
	of "Thu, 13 May 2004 12:17:24 +1000")
References: <BAY12-F29e3nJIxA0hk00012c7d@hotmail.com>
Message-ID: <85y8nxkohs.fsf@servant.blindglobe.net>

"Scott Bray" <scottbray83 at hotmail.com> writes:

> i am working on a statistics type project for university study. i have
> a set of data, have built a discrete probability distribution from
> this data (using the cauchy distribution) and now want to normalize
> it. currently, the area of the distribution is not equal to one. i
> have been trying to find literature about how to normalize, but have
> been unsuccessful (and what i have found, i am unsure on the
> validity). some say to normalize the data points by:
>
> (data point - sample mean) / sample std

This is to "normalize" it in the sense of making it match a N(0,1)
distribution -- you explicitly said that you were using a cauchy as
the basis?!

> others say to multiply by a normalising constant that is "chosen" to
> make the area equal to one. i tried this by just scaling the area to
> equal one.

This is probably the most reasonable approach to making a
"distribution" of your discretized empirical observations.

best,
-tony

-- 
rossini at u.washington.edu            http://www.analytics.washington.edu/ 
Biomedical and Health Informatics   University of Washington
Biostatistics, SCHARP/HVTN          Fred Hutchinson Cancer Research Center
UW (Tu/Th/F): 206-616-7630 FAX=206-543-3461 | Voicemail is unreliable
FHCRC  (M/W): 206-667-7025 FAX=206-667-4812 | use Email

CONFIDENTIALITY NOTICE: This e-mail message and any attachments may be
confidential and privileged. If you received this message in error,
please destroy it and notify the sender. Thank you.



From stefan_hd2004 at yahoo.de  Thu May 13 02:59:52 2004
From: stefan_hd2004 at yahoo.de (=?iso-8859-1?q?Stefan=20Stefan?=)
Date: Thu, 13 May 2004 08:59:52 +0200 (CEST)
Subject: [SciPy-user] problem with scipy installation on Athlon 64 bit system
Message-ID: <20040513065952.43647.qmail@web25207.mail.ukl.yahoo.com>

Can anyone help me to find out why my installation of 
Scipy does not work
I had no problems installing it on my notebook, but
cannot install it on my desktop computer with an
Athlon 64 bit. I tried to follow all instructions but
could not compile.
I think that I installed atlas and f2py correctly.
However, the
python setup.py install for python let to errors which
I list up at the end of this message. I fully 
understand it if you cannot provide any support. If
this is the case could you tell me where I can obtain
additional information on this topic?
It seems something went wrong with the libraries. I
use the Suse 9.1 Linux distribution.

 Best regards,
 Stefan
>

>
> build/temp.linux-x86_64-2.3/config_pygist/config.h
> exist.
> Skipping pygist configuration (remove
build/temp.linux-x86_64-2.3/config_pygist/Make.cfg to
force reconfiguration).
>
**********************************************************************
> x11_info:
>  FOUND:
>    libraries = ['X11']
>    library_dirs = ['/usr/X11R6/lib']
>    include_dirs = ['/usr/X11R6/include']
>
> fftw_info:
>  NOT AVAILABLE
>
> dfftw_info:
>  NOT AVAILABLE
>
>
>    FFTW (http://www.fftw.org/) libraries not found.
>    Directories to search for the libraries can be
specified in the
>    scipy_distutils/site.cfg file (section [fftw]) or
by setting
>    the FFTW environment variable.
> djbfft_info:
>  NOT AVAILABLE
>
>
>    DJBFFT (http://cr.yp.to/djbfft.html) libraries
not found.
>    Directories to search for the libraries can be
specified in the
>    scipy_distutils/site.cfg file (section [djbfft])
or by setting
>    the DJBFFT environment variable.
> lapack_opt_info:
> atlas_threads_info:
> scipy_distutils.system_info.atlas_threads_info
>  NOT AVAILABLE
>
> atlas_info:
> scipy_distutils.system_info.atlas_info
>  FOUND:
>    libraries = ['lapack', 'f77blas', 'cblas',
'atlas']
>    library_dirs = ['/usr/local/lib/atlas']
>    language = f77
>    include_dirs = ['/usr/local/lib/atlas']
>
> running build_src
> building extension "atlas_version" sources
>  adding
'build/src/atlas_version_0xa900e03cf7166de.c' to
sources.
> running build_ext
> customize UnixCCompiler
> customize UnixCCompiler using build_ext
> customize GnuFCompiler
> customize GnuFCompiler
> customize GnuFCompiler using build_ext
> building 'atlas_version' extension
> compling C sources
> gcc options: '-pthread -fno-strict-aliasing -DNDEBUG
-DHAVE_LARGEFILE_SUPPORT -O2 -fmessage-length=0 -Wall
-fPIC'
> compile options: '-I/usr/local/lib/atlas
-I/usr/include/python2.3 -c'
> /usr/bin/g77 -shared
build/temp.linux-x86_64-2.3/build/src/atlas_version_0xa900e03cf7166de.o
-L/usr/local/lib/atlas -latlas -lg2c -o
build/lib.linux-x86_64-2.3/atlas_version.so
>
/usr/lib64/gcc-lib/x86_64-suse-linux/3.3.3/../../../../x86_64-suse-linux/bin/ld:
/usr/local/lib/atlas/libatlas.a(ATL_buildinfo.o):
relocation R_X86_64_32 can not be used when making a
shared object; recompile with -fPIC
> /usr/local/lib/atlas/libatlas.a: could not read
symbols: Bad value
> collect2: ld returned 1 exit status
>
/usr/lib64/gcc-lib/x86_64-suse-linux/3.3.3/../../../../x86_64-suse-linux/bin/ld:
/usr/local/lib/atlas/libatlas.a(ATL_buildinfo.o):
relocation R_X86_64_32 can not be used when making a
shared object; recompile with -fPIC
> /usr/local/lib/atlas/libatlas.a: could not read
symbols: Bad value
> collect2: ld returned 1 exit status
> ##### msg: error: Command "/usr/bin/g77 -shared
build/temp.linux-x86_64-2.3/build/src/atlas_version_0xa900e03cf7166de.o
-L/usr/local/lib/atlas -latlas -lg2c -o
build/lib.linux-x86_64-2.3/atlas_version.so" failed
with exit status 1
> error: Command "/usr/bin/g77 -shared
build/temp.linux-x86_64-2.3/build/src/atlas_version_0xa900e03cf7166de.o
-L/usr/local/lib/atlas -latlas -lg2c -o
build/lib.linux-x86_64-2.3/atlas_version.so" failed
with exit status 1
>  FOUND:
>    libraries = ['lapack', 'f77blas', 'cblas',
'atlas']
>    library_dirs = ['/usr/local/lib/atlas']
>    language = f77
>    define_macros = [('NO_ATLAS_INFO', 2)]
>    include_dirs = ['/usr/local/lib/atlas']
>
> blas_opt_info:
> atlas_blas_threads_info:
> scipy_distutils.system_info.atlas_blas_threads_info
>  NOT AVAILABLE
>
> atlas_blas_info:
> scipy_distutils.system_info.atlas_blas_info
>  FOUND:
>    libraries = ['f77blas', 'cblas', 'atlas']
>    library_dirs = ['/usr/local/lib/atlas']
>    language = c
>    include_dirs = ['/usr/local/lib/atlas']
>
> running build_src
> building extension "atlas_version" sources
>  adding
'build/src/atlas_version_0x78c19bb379372753.c' to
sources.
> running build_ext
> customize UnixCCompiler
> customize UnixCCompiler using build_ext
> building 'atlas_version' extension
> compling C sources
> gcc options: '-pthread -fno-strict-aliasing -DNDEBUG
-DHAVE_LARGEFILE_SUPPORT -O2 -fmessage-length=0 -Wall
-fPIC'
> compile options: '-I/usr/local/lib/atlas
-I/usr/include/python2.3 -c'
> gcc -pthread -shared
build/temp.linux-x86_64-2.3/build/src/atlas_version_0x78c19bb379372753.o
-L/usr/local/lib/atlas -latlas -o
build/lib.linux-x86_64-2.3/atlas_version.so
>
/usr/lib64/gcc-lib/x86_64-suse-linux/3.3.3/../../../../x86_64-suse-linux/bin/ld:
/usr/local/lib/atlas/libatlas.a(ATL_buildinfo.o):
relocation R_X86_64_32 can not be used when making a
shared object; recompile with -fPIC
> /usr/local/lib/atlas/libatlas.a: could not read
symbols: Bad value
> collect2: ld returned 1 exit status
>
/usr/lib64/gcc-lib/x86_64-suse-linux/3.3.3/../../../../x86_64-suse-linux/bin/ld:
/usr/local/lib/atlas/libatlas.a(ATL_buildinfo.o):
relocation R_X86_64_32 can not be used when making a
shared object; recompile with -fPIC
> /usr/local/lib/atlas/libatlas.a: could not read
symbols: Bad value
> collect2: ld returned 1 exit status
> ##### msg: error: Command "gcc -pthread -shared
build/temp.linux-x86_64-2.3/build/src/atlas_version_0x78c19bb379372753.o
-L/usr/local/lib/atlas -latlas -o
build/lib.linux-x86_64-2.3/atlas_version.so" failed
with exit status 1
> error: Command "gcc -pthread -shared
build/temp.linux-x86_64-2.3/build/src/atlas_version_0x78c19bb379372753.o
-L/usr/local/lib/atlas -latlas -o
build/lib.linux-x86_64-2.3/atlas_version.so" failed
with exit status 1
>  FOUND:
>    libraries = ['f77blas', 'cblas', 'atlas']
>    library_dirs = ['/usr/local/lib/atlas']
>    language = c
>    define_macros = [('NO_ATLAS_INFO', 2)]
>    include_dirs = ['/usr/local/lib/atlas']
>
> numpy_info:
>  FOUND:
>    define_macros = [('NUMERIC_VERSION',
'"\\"23.1\\""')]
>    include_dirs = ['/usr/include/python2.3']
>
> SciPy Version 0.3
>
****************************************************************
>    Using fortran_libraries setup option is
depreciated
>   
---------------------------------------------------
>    Use libraries option instead. Yes,
scipy_distutils
>    now supports Fortran sources in libraries.
>
****************************************************************
> running install
> running build
> running config_fc
> running build_src
> building extension "scipy.io.numpyio" sources
> building extension "scipy.xxx.spam" sources
>  adding
'build/temp.linux-x86_64-2.3/Lib/xxx/spam.pyf' to
sources.
>  adding 'build/src/fortranobject.c' to sources.
>  adding 'build/src' to include_dirs.
> building extension "scipy.cluster._vq" sources
> building extension "scipy.xplt.gistC" sources
>  adding 'Lib/xplt/src/play/unix/dir.c' to sources.
>  adding 'Lib/xplt/src/play/unix/files.c' to sources.
>  adding 'Lib/xplt/src/play/unix/fpuset.c' to
sources.
>  adding 'Lib/xplt/src/play/unix/pathnm.c' to
sources.
>  adding 'Lib/xplt/src/play/unix/timew.c' to sources.
>  adding 'Lib/xplt/src/play/unix/uevent.c' to
sources.
>  adding 'Lib/xplt/src/play/unix/ugetc.c' to sources.
>  adding 'Lib/xplt/src/play/unix/umain.c' to sources.
>  adding 'Lib/xplt/src/play/unix/usernm.c' to
sources.
>  adding 'Lib/xplt/src/play/unix/slinks.c' to
sources.
>  adding 'Lib/xplt/src/play/x11/colors.c' to sources.
>  adding 'Lib/xplt/src/play/x11/connect.c' to
sources.
>  adding 'Lib/xplt/src/play/x11/cursors.c' to
sources.
>  adding 'Lib/xplt/src/play/x11/errors.c' to sources.
>  adding 'Lib/xplt/src/play/x11/events.c' to sources.
>  adding 'Lib/xplt/src/play/x11/fills.c' to sources.
>  adding 'Lib/xplt/src/play/x11/fonts.c' to sources.
>  adding 'Lib/xplt/src/play/x11/images.c' to sources.
>  adding 'Lib/xplt/src/play/x11/lines.c' to sources.
>  adding 'Lib/xplt/src/play/x11/pals.c' to sources.
>  adding 'Lib/xplt/src/play/x11/pwin.c' to sources.
>  adding 'Lib/xplt/src/play/x11/resource.c' to
sources.
>  adding 'Lib/xplt/src/play/x11/rgbread.c' to
sources.
>  adding 'Lib/xplt/src/play/x11/textout.c' to
sources.
>  adding 'Lib/xplt/src/play/x11/rect.c' to sources.
>  adding 'Lib/xplt/src/play/x11/clips.c' to sources.
>  adding 'Lib/xplt/src/play/x11/points.c' to sources.
>  adding 'Lib/xplt/src/play/all/hash.c' to sources.
>  adding 'Lib/xplt/src/play/all/hash0.c' to sources.
>  adding 'Lib/xplt/src/play/all/mm.c' to sources.
>  adding 'Lib/xplt/src/play/all/alarms.c' to sources.
>  adding 'Lib/xplt/src/play/all/pstrcpy.c' to
sources.
>  adding 'Lib/xplt/src/play/all/pstrncat.c' to
sources.
>  adding 'Lib/xplt/src/play/all/p595.c' to sources.
>  adding 'Lib/xplt/src/play/all/bitrev.c' to sources.
>  adding 'Lib/xplt/src/play/all/bitlrot.c' to
sources.
>  adding 'Lib/xplt/src/play/all/bitmrot.c' to
sources.
> building extension "scipy.stats.rand" sources
> building extension "scipy.stats.statlib" sources
>  adding 'build/src/fortranobject.c' to sources.
>  adding 'build/src' to include_dirs.
> building extension "scipy.stats.futil" sources
>  adding 'build/src/fortranobject.c' to sources.
>  adding 'build/src' to include_dirs.
> building extension "scipy.interpolate._fitpack"
sources
> building extension "scipy.interpolate.dfitpack"
sources
>  adding 'build/src/fortranobject.c' to sources.
>  adding 'build/src' to include_dirs.
>  adding
'build/src/Lib/interpolate/dfitpack-f2pywrappers.f' to
sources.
> building extension "scipy.fftpack._fftpack" sources
>  adding 'build/src/fortranobject.c' to sources.
>  adding 'build/src' to include_dirs.
> building extension "scipy.fftpack.convolve" sources
>  adding 'build/src/fortranobject.c' to sources.
>  adding 'build/src' to include_dirs.
> building extension "scipy.special.cephes" sources
> building extension "scipy.special.specfun" sources
>  adding 'build/src/fortranobject.c' to sources.
>  adding 'build/src' to include_dirs.
> building extension "scipy.linalg.fblas" sources
>  adding 'build/src/fblas.pyf' to sources.
>  adding 'build/src/fortranobject.c' to sources.
>  adding 'build/src' to include_dirs.
> building extension "scipy.linalg.cblas" sources
>  adding 'build/src/cblas.pyf' to sources.
>  adding 'build/src/fortranobject.c' to sources.
> building extension "scipy.linalg.flapack" sources
>  adding 'build/src/flapack.pyf' to sources.
>  adding 'build/src/fortranobject.c' to sources.
> building extension "scipy.linalg.clapack" sources
>  adding 'build/src/clapack.pyf' to sources.
>  adding 'build/src/fortranobject.c' to sources.
> building extension "scipy.linalg._flinalg" sources
>  adding 'build/src/fortranobject.c' to sources.
> building extension "scipy.linalg.calc_lwork" sources
>  adding 'build/src/fortranobject.c' to sources.
> building extension "scipy.linalg.atlas_version"
sources
> building extension "scipy.signal.sigtools" sources
> building extension "scipy.signal.spline" sources
> building extension "scipy.sparse._zsuperlu" sources
> building extension "scipy.sparse._dsuperlu" sources
> building extension "scipy.sparse._csuperlu" sources
> building extension "scipy.sparse._ssuperlu" sources
> building extension "scipy.sparse._sparsekit" sources
>  adding 'build/src/fortranobject.c' to sources.
>  adding
'build/src/Lib/sparse/_sparsekit-f2pywrappers.f' to
sources.
> building extension "scipy.optimize._minpack" sources
> building extension "scipy.optimize._zeros" sources
> building extension "_lbfgsb" sources
>  adding 'build/src/fortranobject.c' to sources.
> building extension "moduleTNC" sources
> building extension "scipy.integrate._quadpack"
sources
> building extension "scipy.integrate._odepack"
sources
> building extension "scipy.integrate.vode" sources
>  adding 'build/src/fortranobject.c' to sources.
>  adding 'build/src' to include_dirs.
> building extension "scipy_base.fastumath" sources
>  adding 'scipy_core/scipy_base/fastumathmodule.c' to
sources.
> building extension "scipy_base._compiled_base"
sources
> building library "c_misc" sources
> building library "cephes" sources
> building library "rootfind" sources
> building library "statlib" sources
> building library "fitpack" sources
> building library "dfftpack" sources
> building library "mach" sources
> building library "amos" sources
> building library "toms" sources
> building library "cdf" sources
> building library "specfun" sources
> building library "minpack" sources
> building library "superlu_src" sources
> building library "sparsekit_src" sources
> building library "quadpack_src" sources
> building library "linpack_lite_src" sources
> building library "mach_src" sources
> building library "odepack_src" sources
> running build_py
> package init file 'Lib/tests/__init__.py' not found
(or not a regular file)
> package init file
'/home/jordan/installdirectory/SciPy_complete-0.3/Lib/io/tests/__init__.py'
not found (or not a regular file)
> package init file
'/home/jordan/installdirectory/SciPy_complete-0.3/Lib/xxx/tests/__init__.py'
not found (or not a regular file)
> package init file
'/home/jordan/installdirectory/SciPy_complete-0.3/Lib/xxx/yyy/tests/__init__.py'
not found (or not a regular file)
> package init file
'/home/jordan/installdirectory/SciPy_complete-0.3/Lib/cluster/tests/__init__.py'
not found (or not a regular file)
> package init file
'/home/jordan/installdirectory/SciPy_complete-0.3/Lib/gui_thread/tests/__init__.py'
not found (or not a regular file)
> package init file
'/home/jordan/installdirectory/SciPy_complete-0.3/Lib/stats/tests/__init__.py'
not found (or not a regular file)
> package init file
'/home/jordan/installdirectory/SciPy_complete-0.3/Lib/interpolate/tests/__init__.py'
not found (or not a regular file)
> package init file
'/home/jordan/installdirectory/SciPy_complete-0.3/Lib/fftpack/tests/__init__.py'
not found (or not a regular file)
> package init file
'/home/jordan/installdirectory/SciPy_complete-0.3/Lib/fftpack/tests/__init__.py'
not found (or not a regular file)
> package init file
'/home/jordan/installdirectory/SciPy_complete-0.3/Lib/special/tests/__init__.py'
not found (or not a regular file)
> package init file
'/home/jordan/installdirectory/SciPy_complete-0.3/Lib/linalg/tests/__init__.py'
not found (or not a regular file)
> package init file
'/home/jordan/installdirectory/SciPy_complete-0.3/Lib/signal/tests/__init__.py'
not found (or not a regular file)
> package init file
'/home/jordan/installdirectory/SciPy_complete-0.3/Lib/optimize/tests/__init__.py'
not found (or not a regular file)
> package init file
'scipy_core/weave/tests/__init__.py' not found (or not
a regular file)
> package init file
'scipy_core/weave/tests/__init__.py' not found (or not
a regular file)
> package init file
'/home/jordan/installdirectory/SciPy_complete-0.3/scipy_core/scipy_base/tests/__init__.py'
not found (or not a regular file)
> package init file 'Lib/tests/__init__.py' not found
(or not a regular file)
> package init file
'/home/jordan/installdirectory/SciPy_complete-0.3/Lib/io/tests/__init__.py'
not found (or not a regular file)
> package init file
'/home/jordan/installdirectory/SciPy_complete-0.3/Lib/xxx/tests/__init__.py'
not found (or not a regular file)
> package init file
'/home/jordan/installdirectory/SciPy_complete-0.3/Lib/xxx/yyy/tests/__init__.py'
not found (or not a regular file)
> package init file
'/home/jordan/installdirectory/SciPy_complete-0.3/Lib/cluster/tests/__init__.py'
not found (or not a regular file)
> package init file
'/home/jordan/installdirectory/SciPy_complete-0.3/Lib/gui_thread/tests/__init__.py'
not found (or not a regular file)
> package init file
'/home/jordan/installdirectory/SciPy_complete-0.3/Lib/stats/tests/__init__.py'
not found (or not a regular file)
> package init file
'/home/jordan/installdirectory/SciPy_complete-0.3/Lib/interpolate/tests/__init__.py'
not found (or not a regular file)
> package init file
'/home/jordan/installdirectory/SciPy_complete-0.3/Lib/fftpack/tests/__init__.py'
not found (or not a regular file)
> package init file
'/home/jordan/installdirectory/SciPy_complete-0.3/Lib/fftpack/tests/__init__.py'
not found (or not a regular file)
> package init file
'/home/jordan/installdirectory/SciPy_complete-0.3/Lib/special/tests/__init__.py'
not found (or not a regular file)
> package init file
'/home/jordan/installdirectory/SciPy_complete-0.3/Lib/linalg/tests/__init__.py'
not found (or not a regular file)
> package init file
'/home/jordan/installdirectory/SciPy_complete-0.3/Lib/signal/tests/__init__.py'
not found (or not a regular file)
> package init file
'/home/jordan/installdirectory/SciPy_complete-0.3/Lib/optimize/tests/__init__.py'
not found (or not a regular file)
> package init file
'scipy_core/weave/tests/__init__.py' not found (or not
a regular file)
> package init file
'scipy_core/weave/tests/__init__.py' not found (or not
a regular file)
> package init file
'/home/jordan/installdirectory/SciPy_complete-0.3/scipy_core/scipy_base/tests/__init__.py'
not found (or not a regular file)
> running build_clib
> customize UnixCCompiler
> customize UnixCCompiler using build_clib
> customize GnuFCompiler
> customize GnuFCompiler
> customize GnuFCompiler using build_clib
> running build_ext
> customize UnixCCompiler
> customize UnixCCompiler using build_ext
> customize GnuFCompiler
> customize GnuFCompiler
> customize GnuFCompiler using build_ext
> building 'scipy.xplt.gistC' extension
> compling C sources
> gcc options: '-pthread -fno-strict-aliasing -DNDEBUG
-DHAVE_LARGEFILE_SUPPORT -O2 -fmessage-length=0 -Wall
-fPIC'
> compile options:
'-Ibuild/temp.linux-x86_64-2.3/config_pygist
-ILib/xplt/src/gist -ILib/xplt/src/play
-ILib/xplt/src/play/unix -I/usr/X11R6/include
-I/usr/X11R6/include -I/usr/include/python2.3 -c'
> extra options:
'-DGISTPATH="\"/usr/lib64/python2.3/site-packages/scipy/xplt/gistdata\""'
> gcc -pthread -shared
build/temp.linux-x86_64-2.3/Lib/xplt/pygist/gistCmodule.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/gist/gist.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/gist/tick.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/gist/tick60.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/gist/engine.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/gist/gtext.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/gist/draw.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/gist/draw0.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/gist/clip.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/gist/gread.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/gist/gcntr.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/gist/hlevel.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/gist/ps.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/gist/cgm.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/gist/eps.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/gist/style.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/gist/xfancy.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/gist/xbasic.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/unix/dir.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/unix/files.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/unix/fpuset.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/unix/pathnm.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/unix/timew.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/unix/uevent.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/unix/ugetc.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/unix/umain.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/unix/usernm.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/unix/slinks.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/x11/colors.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/x11/connect.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/x11/cursors.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/x11/errors.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/x11/events.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/x11/fills.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/x11/fonts.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/x11/images.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/x11/lines.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/x11/pals.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/x11/pwin.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/x11/resource.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/x11/rgbread.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/x11/textout.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/x11/rect.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/x11/clips.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/x11/points.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/all/hash.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/all/hash0.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/all/mm.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/all/alarms.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/all/pstrcpy.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/all/pstrncat.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/all/p595.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/all/bitrev.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/all/bitlrot.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/all/bitmrot.o
-LLib/xplt/. -LLib/xplt/src -L/usr/X11R6/lib
-L/usr/lib -L/usr/X11R6/lib
-Lbuild/temp.linux-x86_64-2.3 -lX11 -lm -o
build/lib.linux-x86_64-2.3/scipy/xplt/gistC.so
>
/usr/lib64/gcc-lib/x86_64-suse-linux/3.3.3/../../../../x86_64-suse-linux/bin/ld:
skipping incompatible /usr/X11R6/lib/libX11.so when
searching for -lX11
>
/usr/lib64/gcc-lib/x86_64-suse-linux/3.3.3/../../../../x86_64-suse-linux/bin/ld:
skipping incompatible /usr/X11R6/lib/libX11.a when
searching for -lX11
>
/usr/lib64/gcc-lib/x86_64-suse-linux/3.3.3/../../../../x86_64-suse-linux/bin/ld:
skipping incompatible /usr/X11R6/lib/libX11.so when
searching for -lX11
>
/usr/lib64/gcc-lib/x86_64-suse-linux/3.3.3/../../../../x86_64-suse-linux/bin/ld:
skipping incompatible /usr/X11R6/lib/libX11.a when
searching for -lX11
>
/usr/lib64/gcc-lib/x86_64-suse-linux/3.3.3/../../../../x86_64-suse-linux/bin/ld:
cannot find -lX11
> collect2: ld returned 1 exit status
>
/usr/lib64/gcc-lib/x86_64-suse-linux/3.3.3/../../../../x86_64-suse-linux/bin/ld:
skipping incompatible /usr/X11R6/lib/libX11.so when
searching for -lX11
>
/usr/lib64/gcc-lib/x86_64-suse-linux/3.3.3/../../../../x86_64-suse-linux/bin/ld:
skipping incompatible /usr/X11R6/lib/libX11.a when
searching for -lX11
>
/usr/lib64/gcc-lib/x86_64-suse-linux/3.3.3/../../../../x86_64-suse-linux/bin/ld:
skipping incompatible /usr/X11R6/lib/libX11.so when
searching for -lX11
>
/usr/lib64/gcc-lib/x86_64-suse-linux/3.3.3/../../../../x86_64-suse-linux/bin/ld:
skipping incompatible /usr/X11R6/lib/libX11.a when
searching for -lX11
>
/usr/lib64/gcc-lib/x86_64-suse-linux/3.3.3/../../../../x86_64-suse-linux/bin/ld:
cannot find -lX11
> collect2: ld returned 1 exit status
> error: Command "gcc -pthread -shared
build/temp.linux-x86_64-2.3/Lib/xplt/pygist/gistCmodule.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/gist/gist.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/gist/tick.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/gist/tick60.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/gist/engine.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/gist/gtext.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/gist/draw.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/gist/draw0.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/gist/clip.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/gist/gread.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/gist/gcntr.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/gist/hlevel.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/gist/ps.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/gist/cgm.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/gist/eps.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/gist/style.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/gist/xfancy.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/gist/xbasic.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/unix/dir.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/unix/files.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/unix/fpuset.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/unix/pathnm.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/unix/timew.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/unix/uevent.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/unix/ugetc.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/unix/umain.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/unix/usernm.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/unix/slinks.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/x11/colors.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/x11/connect.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/x11/cursors.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/x11/errors.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/x11/events.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/x11/fills.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/x11/fonts.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/x11/images.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/x11/lines.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/x11/pals.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/x11/pwin.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/x11/resource.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/x11/rgbread.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/x11/textout.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/x11/rect.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/x11/clips.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/x11/points.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/all/hash.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/all/hash0.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/all/mm.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/all/alarms.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/all/pstrcpy.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/all/pstrncat.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/all/p595.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/all/bitrev.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/all/bitlrot.o
build/temp.linux-x86_64-2.3/Lib/xplt/src/play/all/bitmrot.o
-LLib/xplt/. -LLib/xplt/src -L/usr/X11R6/lib
-L/usr/lib -L/usr/X11R6/lib
-Lbuild/temp.linux-x86_64-2.3 -lX11 -lm -o
build/lib.linux-x86_64-2.3/scipy/xplt/gistC.so" failed
with exit status 1
>
>
> linux:/home/xxxx/installdirectory/SciPy_complete-0.3
#
>




	

	
		
Mit sch?nen Gr??en von Yahoo! Mail - http://mail.yahoo.de



From jl at dmi.dk  Thu May 13 06:51:24 2004
From: jl at dmi.dk (Jesper Larsen)
Date: Thu, 13 May 2004 12:51:24 +0200
Subject: [SciPy-user] High-pass filtering
In-Reply-To: <40A28F6A.6020901@ee.byu.edu>
References: <200405121614.27707.jl@dmi.dk> <40A28F6A.6020901@ee.byu.edu>
Message-ID: <200405131251.24858.jl@dmi.dk>

Thanks for the help Travis,

> SciPy uses Numeric and doesn't require numarray (it can use numarray
> objects as inputs which will be treated as Python sequences and can be
> slow).

Okay, I did'nt know that (I'm quite new to Python and this is my first use of 
SciPy). I thought that numarray was a replacement for Numeric and have 
therefore only bothered to learn the former (maybe I should have a look at 
Numeric as well).

> What kind of high-pass filter are you looking for.  Simple FIR filters
> are quite simple to design and give linear phase.  Other IIR filters are
> available.  I'll show you how to design a high-pass FIR filter using the
> remez algorithm (you could also look at signal.firwin for windowed
> filter design (but remez is usually a better solution).
...

I think the FIR filter you have shown how to use is just fine for my purposes 
(but I am obviously not an expert on filters). I have therefore described 
below what I am going to use it for. If you have any ideas of a better filter 
for this application I would be grateful to hear about it.

I'm making an application for quality control (QC) of timeseries of 
observations from various meteorological stations. The QC first has several 
simple checks. After these checks a check based on optimal interpolation (OI) 
is performed. OI is a method to estimate a given observation at a given 
station using statistical covariances with timeseries from other stations 
(including these timeseries with lagged observations). The OI method will 
also give an estimate of the uncertainty of the estimated value. If the real 
value is to far away from the estimated value it will be removed or replaced 
by the estimated value.

To get good results from the OI it is necessary to remove low frequency 
oscillations (timescales > 2-3 months). For example the annual oscillation in 
temperature should be removed. If this is not done the covariances used in 
the OI will be overshadowed by the annual oscillation and it will be 
impossible to estimate fluctuations with high frequency (timescale of 
hours-days). The reason for this is that the covariance structure of low and 
high frequency oscillations are not the same due to the fact that different 
physical processes are responsible for each type of variability.

I hope to release the application as open source when it is complete and has 
been tested (it still requires quite some work). It might then be 
adapted/expanded to QC of other types of observational timeseries in 
completely different areas.

Kind regards,
Jesper



From JADDISON at SYSTEMS.TEXTRON.com  Sun May 16 23:29:30 2004
From: JADDISON at SYSTEMS.TEXTRON.com (Addison, Jason)
Date: Sun, 16 May 2004 17:29:30 -1000
Subject: [SciPy-user] pilutil toimage
Message-ID: <200405161729.30692.jaddison@systems.textron.com>

I'm trying to generate an image from a 2D array and a colormap.

Here is a simple example:

# python code <><><><><><><><><><><><><><><>

c0 = array([[0, 0, 0, 0], [1, 1, 1, 1], [2, 2, 2, 2]])

icm = array([[1, 0, 0], [0, 1, 0], [0, 0, 1]])

c0_im = toimage(c0, pal=icm)

c0_im.save('c0.png')

# python code <><><><><><><><><><><><><><><>
# Fedora Core 1
# python version 2.3.3
# scipy version 0.3
# Numeric 23.1
# PIL 1.1.4

But, I always get a grey scale image. I think that I should 
get 3 pixel by 4 pixel with the first row red, then green, 
then blue, but I get black, grey and white.

What am I missing? 

Does anyone have some examples of turning 2D arrays into 
color graphics files?

Thank you,
Jason



From eric at enthought.com  Mon May 17 00:34:59 2004
From: eric at enthought.com (eric jones)
Date: Sun, 16 May 2004 23:34:59 -0500
Subject: [SciPy-user] ANN: Job Openings at Enthought
Message-ID: <40A840F3.2050608@enthought.com>

Hello,

This is a bit of a status report I guess for SciPy (www.scipy.org) and
Enthought (www.enthought.com), but, really, it is a long winded job
posting. 

Among other things, Enthought develops scientific software
for clients. Python is central to to our strategy in this arena.  We have
long supported SciPy and believe strongly in Open Source Software. 
The purpose is to give people a feeling about what we do, the technical
challenges we face, and the skills needed to meet them.

SciPy
-----
There has been a lot of work on SciPy lately by the Travis Oliphant, Travis
Vaught, Pearu Peterson, Joe Cooper, and on the website by Jon-Eric 
Steinbomer
and Janet Swisher.  The site is now upgraded to Plone and we have a 
SciPy 0.3
package out.  If you're interested in seeing the activity level at 
scipy.org,
please visit.

http://www.scipy.org/map?rmurl=http%3A//scipy.net/scipyusage/

It looks like we're averaging about a 100 downloads a day if you add up all
source and binary distributions.  When SciPy 0.1 was released a couple a 
years
ago, I think we averaged about 10.  Its extremely difficult to 
extrapolate a user
base from such information, the obvious growth is good news.

Other Stuff
-----------
In addition to SciPy we have multiple projects going on right now that 
either
already plug into or will be plugins to a Python-based scientific 
application
framework called Envisage (sorta like an IDE for science apps) that we're
working on.  The idea is:

(1) To lower the bar as far as work required to get a GUI put
   on the front end of a scientific algorithm.
(2) Develop scientific plugins that can play well together so
   that, for instance, an electromagnetics simulator built by one
   person can be used in conjunction with an optimization plugin
   built by another person and results can be visualized with a
   3D plugin.

This is a hard, long path that requires much work and though.  We have
started the process and actually have built an app on a very early 
version of
the Envisage framework.  The app works great, but it's usage of the 
framework is a
little mixed at this point.  There were a lot of compromises we ended up 
making
on the framework side in order to ship on time.

Also, I am not sure whether "easy-to-program" and "flexible architecture for
compatible plugins" are not mutually exclusive.  I always have in my 
mind that I
want a smart scientist to be able to build the GUI in a short amount of time
after the algorithm (which is the "hard part") is finished.  I still 
harbor this
wish, but the lessons I've had over the last couple of years suggest 
that the GUI is
actually the most expensive part of development for large apps.  This is 
partially
due to the fact that commercial apps are rarely built around "research" 
algorithms
-- meaning that the algorithm code usually already exists in some form.  
Still,
UIs take tons of time.  Testing them is hard.  Flexibility requires more
complexity (factories, adapters, etc.) than seems necessary at first.  
Further,
building *good* UI's is really hard.  Scientist rarely have the patience or
perspective for it -- and yet it is very often difficult to build a good 
UI for
science apps without a scientist's understanding of the underlying problem.

We have a number of tools along with SciPy that we're building that are 
pieces
of this puzzle.

  1. Traits -- They at the heart of everything we do.  They provide
     some explicit typing, default GUI's, an observer event model for
     anything that uses them.  They can also require some getting used
     to.
  2. Kiva/Enable -- This is the generic drawing system for Python.
  3. Chaco -- Interactive 2D Plotting
  4. PyFace -- MVC layer on top of wxPython.  Supports trees, menus,
     and a few other things right now.  This will grow.
  5. Envisage -- Plugin based GUI framework for scientific applications.
     Beyond the basic GUI that comes from Envisage, we already have a couple
     plugins for profiling code using hotshot and also searching for memory
     leaks.  David Morrill wrote these and they are called gotcha and 
enroll.
     Martin Chilvers wrote a simple scintilla based plugin for editing 
scripts,
     and a PyCrust shell is available by default in the app (not
     a plugin at the moment).
    
     We would love to see a number of other plugins:
      a. Debugger.
      b. IPython-like PyCrust going.
      c. A more full featured script editing plugin.
         (there is a huge list of features here).
      d. Mayavi
      e. etc.

These are all used actively in commercial products that we deliver.  
However,
they are also in various stages of development and will continue to evolve
and grow.  Some are still pretty green.  Portions of these are openly 
available
now.  All five of the listed tool sets will be open at some point this 
summer
when they are cleaned up a bit.

Jobs
----
All of this leads to the fact that we are hiring and looking for smart,
pleasant, communicative people with integrity that are excited about 
building
this sort of stuff.  There is a lot of software architecture to be done that
needs talented designers.  There is UI design/development that needs 
scientists
that care about UIs or UI developers/designers that are really quick at 
grasping
scientific problems.  We also need really strong scientists/scientific
developers to do some of the backend coding for commercial products and also
SciPy development.  If you have a background in geophysics, electromagnetics
(especially multi-pole methods) that is a big plus. For this, a PhD is 
helpful, but
not required.  Parallel computing skills wouldn't hurt.  3D 
visualization a la
VTK/Python is a major need.  So is knowledge about 2D rendering to help 
finish
out Kiva and the Enable toolset. Strong Python skills, Python extension 
writing
knowledge and strong C/C++ skills are a big benefit.  Design/development 
with or
of Java NetBeans or Eclipse-like architecture is great -- or any other 
solid GUI
architecture.  Dedication to writing clean/clear code meant for other 
humans
to read is a requirement.

We have 3 or 4 scientific/python positions to fill over the coming 
months, and
we're looking for candidates that have the best mix of the above 
skills.  You'll
join our existing team of scientist/engineers, computer scientists, 
technical
writers, and an HCI specialist.  We like this mix and feel it is the 
best way to
build this sort of software. If you are interested in working at Enthought,
please send us your resume at jobs at enthought.com.  If not, please send this
posting to the smartest people you know that fit some part of the above 
description
(Python experience not explicitly required).  Salaries are competitive.  
Candidates must
be willing to relocate to Austin, Tx.
     
thanks,
eric jones

PS: There are additional positions listed at 
http://www.enthought.com/careers.htm
for those interested in business application development (not 
necessarily with Python).




From jeanluc.menut at free.fr  Tue May 18 12:28:44 2004
From: jeanluc.menut at free.fr (Jean-Luc Menut)
Date: Tue, 18 May 2004 18:28:44 +0200
Subject: [SciPy-user] scipy on sun/OSF
Message-ID: <40AA39BC.3090704@free.fr>

Hello,

I work on a sun computer under OSF through telnet or ssh. Is it scipy 
working on this kind of computer/OS ? I need to use some plotting 
capability (2D, 3D, images) and calculation like FFT, etc... It will 
replace IDL.

I'd like to have a confirmation thet it is possible before starting to 
install all the software needed by scipy.

best regards,
Jean-Luc



From jeanluc.menut at free.fr  Wed May 19 09:54:47 2004
From: jeanluc.menut at free.fr (Jean-Luc Menut)
Date: Wed, 19 May 2004 15:54:47 +0200
Subject: [SciPy-user] Problem when building scipy 0.3 for alpha OSF1
Message-ID: <40AB6727.6090000@free.fr>

Hello,

I hope i'm writing in the good mailing list.

I have a problem when I build  scipy on my alpha (unix : OSF1):

After the compilation of python 2.2.3, the installation of Numeric 22.0 
and F2PY-2.39.235, i install the binary of atlas 3.4.1 (for OSF1 of 
course). When  i run python setup.py build, there is a lot a compilation 
but after a moment it's stop :

I have this error  message :

g++ -shared -expect_unresolved * 
build/temp.osf1-V5.1-alpha-2.2/vq_wrap.o 
-Lbuild/temp.osf1-V5.1-alpha-2.2 -o 
build/lib.osf1-V5.1-alpha-2.2/scipy/cluster/_vq.so
/usr/bin/ld:
Object file format error in: DEVELOPERS.txt: read_cur_obj_info: bad file 
magic number(0x2e2e)
collect2: ld returned 1 exit status
/usr/bin/ld:
Object file format error in: DEVELOPERS.txt: read_cur_obj_info: bad file 
magic number(0x2e2e)
collect2: ld returned 1 exit status

1) it's look strange because there is a pb with ld and DEVELOPERS.txt 
wich is a text file.
2)I can't try with the gnu ld because it don't work on alpha/OSF1
3) it's the same kind of error with a own compilated version of atlas 
(3.6.0)
4)I don't know if there is a relation but I have a problem with  FFTW 
too. I compile and install it without pb but the building script of 
scipy don't see it.

Someone can help me ?

best regards,
Jean-Luc Menut



From pearu at scipy.org  Wed May 19 10:17:08 2004
From: pearu at scipy.org (Pearu Peterson)
Date: Wed, 19 May 2004 09:17:08 -0500 (CDT)
Subject: [SciPy-user] Problem when building scipy 0.3 for alpha OSF1
In-Reply-To: <40AB6727.6090000@free.fr>
References: <40AB6727.6090000@free.fr>
Message-ID: <Pine.LNX.4.58.0405190912180.17789@scipy.org>



On Wed, 19 May 2004, Jean-Luc Menut wrote:

> Hello,
> 
> I hope i'm writing in the good mailing list.

Yes, that's a good one;-)

> I have a problem when I build  scipy on my alpha (unix : OSF1):
> 
> After the compilation of python 2.2.3, the installation of Numeric 22.0 
> and F2PY-2.39.235, i install the binary of atlas 3.4.1 (for OSF1 of 
> course). When  i run python setup.py build, there is a lot a compilation 
> but after a moment it's stop :
> 
> I have this error  message :
> 
> g++ -shared -expect_unresolved * build/temp.osf1-V5.1-alpha-2.2/vq_wrap.o 
                                ^^^
This 'star' is causing all the trouble: all files in the current directory 
(.txt files etc) are passed to g++. Could you find out where this 
'star' comes from? Can you build other extension modules than of scipy?

Pearu



From jeanluc.menut at free.fr  Wed May 19 10:17:47 2004
From: jeanluc.menut at free.fr (Jean-Luc Menut)
Date: Wed, 19 May 2004 16:17:47 +0200
Subject: [SciPy-user] Problem when building scipy 0.3 for alpha OSF1
In-Reply-To: <Pine.LNX.4.58.0405190912180.17789@scipy.org>
References: <40AB6727.6090000@free.fr>
	<Pine.LNX.4.58.0405190912180.17789@scipy.org>
Message-ID: <40AB6C8B.2030405@free.fr>

>>g++ -shared -expect_unresolved * build/temp.osf1-V5.1-alpha-2.2/vq_wrap.o 
> 
>                                 ^^^
> This 'star' is causing all the trouble: all files in the current directory 
> (.txt files etc) are passed to g++. Could you find out where this 
> 'star' comes from? Can you build other extension modules than of scipy?

Yes, what module by example ? I have built Numeric without problem before.

  Jean-Luc



From pearu at scipy.org  Wed May 19 10:42:12 2004
From: pearu at scipy.org (Pearu Peterson)
Date: Wed, 19 May 2004 09:42:12 -0500 (CDT)
Subject: [SciPy-user] Problem when building scipy 0.3 for alpha OSF1
In-Reply-To: <40AB6C8B.2030405@free.fr>
References: <40AB6727.6090000@free.fr>
	<Pine.LNX.4.58.0405190912180.17789@scipy.org>
	<40AB6C8B.2030405@free.fr>
Message-ID: <Pine.LNX.4.58.0405190939060.17789@scipy.org>



On Wed, 19 May 2004, Jean-Luc Menut wrote:

> >>g++ -shared -expect_unresolved * build/temp.osf1-V5.1-alpha-2.2/vq_wrap.o 
> > 
> >                                 ^^^
> > This 'star' is causing all the trouble: all files in the current directory 
> > (.txt files etc) are passed to g++. Could you find out where this 
> > 'star' comes from? Can you build other extension modules than of scipy?
> 
> Yes, what module by example ? I have built Numeric without problem before.

First try installing scipy_core packages:

  cd scipy_core
  python setup.py install

Then try building scipy modules separately:

  cd Lib/linalg
  python setup_linalg.py build

  cd Lib/fftpack
  python setup_fftpack.py build

  cd Lib/cluster
  python setup_cluster.py build

  etc.

Pearu



From jeanluc.menut at free.fr  Wed May 19 10:46:18 2004
From: jeanluc.menut at free.fr (Jean-Luc Menut)
Date: Wed, 19 May 2004 16:46:18 +0200
Subject: [SciPy-user] Problem when building scipy 0.3 for alpha OSF1
In-Reply-To: <Pine.LNX.4.58.0405190939060.17789@scipy.org>
References: <40AB6727.6090000@free.fr>
	<Pine.LNX.4.58.0405190912180.17789@scipy.org>	<40AB6C8B.2030405@free.fr>
	<Pine.LNX.4.58.0405190939060.17789@scipy.org>
Message-ID: <40AB733A.1050008@free.fr>


> First try installing scipy_core packages:
Ok

>   cd scipy_core
>   python setup.py install

There is no setup.py in scipy_core. I tried to build setup_scipy_base.py 
in scipy_base but it need scipy_distutils and if i try to build 
setup_scipy_distutils.py in scipy_distutils, i have this message :

scipy_distutils Version 0.3.0_27.455
Traceback (most recent call last):
   File "setup_scipy_distutils.py", line 27, in ?
     from core import setup
   File "core.py", line 5, in ?
     from scipy_distutils.dist import Distribution
ImportError: No module named scipy_distutils.dist

thank you,

Jean-Luc



From pearu at scipy.org  Wed May 19 11:05:17 2004
From: pearu at scipy.org (Pearu Peterson)
Date: Wed, 19 May 2004 10:05:17 -0500 (CDT)
Subject: [SciPy-user] Problem when building scipy 0.3 for alpha OSF1
In-Reply-To: <40AB733A.1050008@free.fr>
References: <40AB6727.6090000@free.fr>
	<Pine.LNX.4.58.0405190912180.17789@scipy.org>
	<Pine.LNX.4.58.0405190939060.17789@scipy.org>
	<40AB733A.1050008@free.fr>
Message-ID: <Pine.LNX.4.58.0405191000170.17789@scipy.org>



On Wed, 19 May 2004, Jean-Luc Menut wrote:

> 
> > First try installing scipy_core packages:
> Ok
> 
> >   cd scipy_core
> >   python setup.py install
> 
> There is no setup.py in scipy_core. I tried to build setup_scipy_base.py 
> in scipy_base but it need scipy_distutils and if i try to build 
> setup_scipy_distutils.py in scipy_distutils, i have this message :
> 
> scipy_distutils Version 0.3.0_27.455
> Traceback (most recent call last):
>    File "setup_scipy_distutils.py", line 27, in ?
>      from core import setup
>    File "core.py", line 5, in ?
>      from scipy_distutils.dist import Distribution
> ImportError: No module named scipy_distutils.dist

So, the scipy tar-balls are not complete..

I suggest getting scipy from CVS. See

  http://www.scipy.org/cvs/

(if there will be any bug fixes for your platform then you would need CVS 
version of scipy anyway..)

Or you can add /path/to/scipy_core to PYTHONPATH environment variable and 
then try building scipy_base.

Pearu



From cookedm at physics.mcmaster.ca  Wed May 19 14:19:08 2004
From: cookedm at physics.mcmaster.ca (David M. Cooke)
Date: Wed, 19 May 2004 14:19:08 -0400
Subject: [SciPy-user] Problem when building scipy 0.3 for alpha OSF1
In-Reply-To: <Pine.LNX.4.58.0405190912180.17789@scipy.org>
References: <40AB6727.6090000@free.fr>
	<Pine.LNX.4.58.0405190912180.17789@scipy.org>
Message-ID: <20040519181908.GA11465@arbutus.physics.mcmaster.ca>

On Wed, May 19, 2004 at 09:17:08AM -0500, Pearu Peterson wrote:
> 
> 
> On Wed, 19 May 2004, Jean-Luc Menut wrote:
> 
> > Hello,
> > 
> > I hope i'm writing in the good mailing list.
> 
> Yes, that's a good one;-)
> 
> > I have a problem when I build  scipy on my alpha (unix : OSF1):
> > 
> > After the compilation of python 2.2.3, the installation of Numeric 22.0 
> > and F2PY-2.39.235, i install the binary of atlas 3.4.1 (for OSF1 of 
> > course). When  i run python setup.py build, there is a lot a compilation 
> > but after a moment it's stop :
> > 
> > I have this error  message :
> > 
> > g++ -shared -expect_unresolved * build/temp.osf1-V5.1-alpha-2.2/vq_wrap.o 
>                                 ^^^
> This 'star' is causing all the trouble: all files in the current directory 
> (.txt files etc) are passed to g++. Could you find out where this 
> 'star' comes from? Can you build other extension modules than of scipy?

I've had similiar problems compiling on Tru64. The -expect_unresolved
"*" (note the quotes) comes from Python's Makefile -- that option is
needed for the linker to work properly. The quotes are getting stripped
off somewhere.

I worked around this by writing wrapper scripts for cc (I use
DEC's/Compaq's/HP's compiler), and ld. By setting the CC and LDSHARED
environment variables (after editing
usr/lib/python2.3/distutils/sysconfig.py to allow overriding LDSHARED),
the wrapper scripts could make the right calls.

You'll want something like this:

#!/usr/bin/env python
import sys
import os
command = 'g++ -shared -expect_unresolved "*" ' + ' '.join(sys.argv[1:])
print '>>>', command
w = os.system(command)
sys.exit(os.WEXITSTATUS(w))

(and similiar for ld). Call it cc-wrapper and ldshared-wrapper, and put
them in your path, and call like this
$ CC=cc-wrapper LDSHARED=ldshared-wrapper python ./setup.py build

[I didn't have the time to fix the underlying quote problem, hence the
hack.]

Hope this helps.

-- 
|>|\/|<
/--------------------------------------------------------------------------\
|David M. Cooke                      http://arbutus.physics.mcmaster.ca/dmc/
|cookedm at physics.mcmaster.ca



From pearu at scipy.org  Wed May 19 16:04:08 2004
From: pearu at scipy.org (Pearu Peterson)
Date: Wed, 19 May 2004 15:04:08 -0500 (CDT)
Subject: [SciPy-user] Problem when building scipy 0.3 for alpha OSF1
In-Reply-To: <20040519181908.GA11465@arbutus.physics.mcmaster.ca>
References: <40AB6727.6090000@free.fr>
	<Pine.LNX.4.58.0405190912180.17789@scipy.org>
	<20040519181908.GA11465@arbutus.physics.mcmaster.ca>
Message-ID: <Pine.LNX.4.58.0405191459280.17789@scipy.org>



On Wed, 19 May 2004, David M. Cooke wrote:

> On Wed, May 19, 2004 at 09:17:08AM -0500, Pearu Peterson wrote:

> > > I have a problem when I build  scipy on my alpha (unix : OSF1):
> > > 
> > > After the compilation of python 2.2.3, the installation of Numeric 22.0 
> > > and F2PY-2.39.235, i install the binary of atlas 3.4.1 (for OSF1 of 
> > > course). When  i run python setup.py build, there is a lot a compilation 
> > > but after a moment it's stop :
> > > 
> > > I have this error  message :
> > > 
> > > g++ -shared -expect_unresolved * build/temp.osf1-V5.1-alpha-2.2/vq_wrap.o 
> >                                 ^^^
> > This 'star' is causing all the trouble: all files in the current directory 
> > (.txt files etc) are passed to g++. Could you find out where this 
> > 'star' comes from? Can you build other extension modules than of scipy?
> 
> I've had similiar problems compiling on Tru64. The -expect_unresolved
> "*" (note the quotes) comes from Python's Makefile -- that option is
> needed for the linker to work properly. The quotes are getting stripped
> off somewhere.

Thanks for the hint. The quotes get stripped off in 
distutils.util.split_quoted function. I have implemented a workaround to 
this issue in scipy_distutils. Now the quotes are preserved when a quoted 
word does not contain any white-space.

So, update scipy from cvs and try again. Let us know if the patch works.

HTH,
Pearu



From nwagner at mecha.uni-stuttgart.de  Fri May 21 03:58:38 2004
From: nwagner at mecha.uni-stuttgart.de (Nils Wagner)
Date: Fri, 21 May 2004 09:58:38 +0200
Subject: [SciPy-user] AttributeError: keys
Message-ID: <40ADB6AE.4070501@mecha.uni-stuttgart.de>

Python 2.3.3 (#1, Apr  6 2004, 01:47:39)
[GCC 3.3.3 (SuSE Linux)] on linux2
Type "help", "copyright", "credits" or "license" for more information.
 >>> from scipy import *
 >>> import IPython
 >>> a = zeros((2,2),Float)
 >>> a
array([[ 0.,  0.],
        [ 0.,  0.]])
 >>> who (a)
Traceback (most recent call last):
   File "<stdin>", line 1, in ?
   File "/usr/lib/python2.3/site-packages/scipy/common.py", line 253, in who
     for name in vardict.keys():
AttributeError: keys
 >>>



From pearu at scipy.org  Fri May 21 06:17:38 2004
From: pearu at scipy.org (Pearu Peterson)
Date: Fri, 21 May 2004 05:17:38 -0500 (CDT)
Subject: [SciPy-user] AttributeError: keys
In-Reply-To: <40ADB6AE.4070501@mecha.uni-stuttgart.de>
References: <40ADB6AE.4070501@mecha.uni-stuttgart.de>
Message-ID: <Pine.LNX.4.58.0405210512220.6900@scipy.org>



On Fri, 21 May 2004, Nils Wagner wrote:

> Python 2.3.3 (#1, Apr  6 2004, 01:47:39)
> [GCC 3.3.3 (SuSE Linux)] on linux2
> Type "help", "copyright", "credits" or "license" for more information.
>  >>> from scipy import *
>  >>> import IPython
>  >>> a = zeros((2,2),Float)
>  >>> a
> array([[ 0.,  0.],
>         [ 0.,  0.]])
>  >>> who (a)
> Traceback (most recent call last):
>    File "<stdin>", line 1, in ?
>    File "/usr/lib/python2.3/site-packages/scipy/common.py", line 253, in who
>      for name in vardict.keys():
> AttributeError: keys

Read help(who). So try

  who()
  who({'a',a})

who(a) could never work in Python as it does in Matlab (who('a')).
However, who(a=a) could.

Regards,
Pearu



From Jonathan.Peirce at nottingham.ac.uk  Fri May 21 09:34:28 2004
From: Jonathan.Peirce at nottingham.ac.uk (Jon Peirce)
Date: Fri, 21 May 2004 14:34:28 +0100
Subject: [SciPy-user] distribution.fit() methods
In-Reply-To: <mailman.0.1085147088.18550.scipy-user@scipy.net>
References: <mailman.0.1085147088.18550.scipy-user@scipy.net>
Message-ID: <40AE0564.10706@psychology.nottingham.ac.uk>

Hi there,

I'm trying to fit a weibull function to some data using Scipy. It looks 
like I should be able to use exponweibull_gen.fit but I'm a bit puzzled 
about the arguments. According to the code the arguments  (data, *args, 
**kwds) are needed, where *args seems to contain the shape parameters 
for the fit and **kwds contains the location and scale params.

But if I'm fitting a function, surely I want to give xData and yData and 
*not* shape arguments (or maybe a first guess at those). Does anyone 
know how to use these methods?

Jon

This message has been scanned but we cannot guarantee that it and any
attachments are free from viruses or other damaging content: you are
advised to perform your own checks.  Email communications with the
University of Nottingham may be monitored as permitted by UK legislation.



From jeanluc.menut at free.fr  Fri May 21 10:28:07 2004
From: jeanluc.menut at free.fr (Jean-Luc Menut)
Date: Fri, 21 May 2004 16:28:07 +0200
Subject: [SciPy-user] Problem when building scipy 0.3 for alpha OSF1
In-Reply-To: <Pine.LNX.4.58.0405191459280.17789@scipy.org>
References: <40AB6727.6090000@free.fr>
	<Pine.LNX.4.58.0405190912180.17789@scipy.org>
	<20040519181908.GA11465@arbutus.physics.mcmaster.ca>
	<Pine.LNX.4.58.0405191459280.17789@scipy.org>
Message-ID: <40AE11F7.9010604@free.fr>

hello,

> Thanks for the hint. The quotes get stripped off in 
> distutils.util.split_quoted function. I have implemented a workaround to 
> this issue in scipy_distutils. Now the quotes are preserved when a quoted 
> word does not contain any white-space.
> 
> So, update scipy from cvs and try again. Let us know if the patch works.


I' ve installed newer version of python and numeric (2.3.3 and 23.1).
When i try to build scipy (the last cvs version) i have another error 
message :


error: Command "gcc -pthread -fno-strict-aliasing -DNDEBUG -g -O3 -Wall 
-Wstrict-prototypes -Ibuild/temp.osf1-V5.1-alpha-2.3/config_pygist 
-ILib/xplt/src/gist -ILib/xplt/src/play -ILib/xplt/src/play/unix 
-I/usr/include -I/usr/include 
-I/home2/menut/prg/python2.3/include/python2.3 -c 
Lib/xplt/pygist/gistCmodule.c -o 
build/temp.osf1-V5.1-alpha-2.3/Lib/xplt/pygist/gistCmodule.o 
-DGISTPATH="\"/home2/menut/prg/python2.3/lib/python2.3/site-packages/scipy/xplt/gistdata\"" 
-DNO_EXP10" failed with exit status 1
1.77u 2.16s 1:00 6% 20+27k 210+332io 0pf+0w


I fact i had this message for the cluster module but i've moved its 
directory  to cluster.sav for testing.

best regards,

jean-luc



From pearu at scipy.org  Fri May 21 10:59:10 2004
From: pearu at scipy.org (Pearu Peterson)
Date: Fri, 21 May 2004 09:59:10 -0500 (CDT)
Subject: [SciPy-user] Problem when building scipy 0.3 for alpha OSF1
In-Reply-To: <40AE11F7.9010604@free.fr>
References: <40AB6727.6090000@free.fr>
	<Pine.LNX.4.58.0405190912180.17789@scipy.org>
	<20040519181908.GA11465@arbutus.physics.mcmaster.ca>
	<40AE11F7.9010604@free.fr>
Message-ID: <Pine.LNX.4.58.0405210955010.22671@scipy.org>



On Fri, 21 May 2004, Jean-Luc Menut wrote:

> I' ve installed newer version of python and numeric (2.3.3 and 23.1).
> When i try to build scipy (the last cvs version) i have another error 
> message :
> 
> 
> error: Command "gcc -pthread -fno-strict-aliasing -DNDEBUG -g -O3 -Wall 
> -Wstrict-prototypes -Ibuild/temp.osf1-V5.1-alpha-2.3/config_pygist 
> -ILib/xplt/src/gist -ILib/xplt/src/play -ILib/xplt/src/play/unix 
> -I/usr/include -I/usr/include 
> -I/home2/menut/prg/python2.3/include/python2.3 -c 
> Lib/xplt/pygist/gistCmodule.c -o 
> build/temp.osf1-V5.1-alpha-2.3/Lib/xplt/pygist/gistCmodule.o 
> -DGISTPATH="\"/home2/menut/prg/python2.3/lib/python2.3/site-packages/scipy/xplt/gistdata\"" 
> -DNO_EXP10" failed with exit status 1
> 1.77u 2.16s 1:00 6% 20+27k 210+332io 0pf+0w

It is not a real error message. It just says that the above command line 
failed. The real error message should have been shown earlier.

Anyways, it's looks like xplt is the only scipy module that fails to 
build. You can disable it by adding 'xplt' to ignore_packags list at the 
end of setup.py file, at least you could then install and test scipy.

Pearu



From jeanluc.menut at free.fr  Fri May 21 12:01:40 2004
From: jeanluc.menut at free.fr (Jean-Luc Menut)
Date: Fri, 21 May 2004 18:01:40 +0200
Subject: [SciPy-user] Problem when building scipy 0.3 for alpha OSF1
In-Reply-To: <Pine.LNX.4.58.0405210955010.22671@scipy.org>
References: <40AB6727.6090000@free.fr>
	<Pine.LNX.4.58.0405190912180.17789@scipy.org>
	<20040519181908.GA11465@arbutus.physics.mcmaster.ca>
	<40AE11F7.9010604@free.fr>
	<Pine.LNX.4.58.0405210955010.22671@scipy.org>
Message-ID: <40AE27E4.70205@free.fr>

Pearu Peterson a ?crit :

> Anyways, it's looks like xplt is the only scipy module that fails to 
> build. You can disable it by adding 'xplt' to ignore_packags list at the 
> end of setup.py file, at least you could then install and test scipy.

I have 6 packages which show the same kind of message :'cluster', 
'xplt', 'stats', 'signal', 'sparse', 'optimize' (1). I can build and 
install scipy without these package but when I write "import scipy" 
under python i have this message :

Python 2.3.3 (#8, May 21 2004, 12:13:46)
[GCC 3.3.3] on osf1V5
Type "help", "copyright", "credits" or "license" for more information.
 >>> import scipy
Floating point exception

and i return on the shell.

Jean-Luc

(1): I copy here the error messages for cluster only :


c++ options: '-pthread -fno-strict-aliasing -DNDEBUG -g -O3 -Wall 
-Wstrict-prototypes'
creating build/temp.osf1-V5.1-alpha-2.3/Lib/cluster
creating build/temp.osf1-V5.1-alpha-2.3/Lib/cluster/src
compile options: '-I/home2/menut/prg/python2.3/include/python2.3 -c'
c++: Lib/cluster/src/vq_wrap.cpp
In file included from 
/home2/menut/prg/python2.3/include/python2.3/Python.h:8,
                  from Lib/cluster/src/vq_wrap.cpp:176:
/home2/menut/prg/python2.3/include/python2.3/pyconfig.h:840:1: warning: 
"_OSF_SOURCE" redefined
In file included from 
/usr/local/lib/gcc-lib/alphaev67-dec-osf5.1/3.3.3/include/string.h:61,
                  from Lib/cluster/src/vq_wrap.cpp:27:
/usr/include/standards.h:206:1: warning: this is the location of the 
previous definition
In file included from 
/home2/menut/prg/python2.3/include/python2.3/Python.h:8,
                  from Lib/cluster/src/vq_wrap.cpp:176:
/home2/menut/prg/python2.3/include/python2.3/pyconfig.h:847:1: warning: 
"_POSIX_C_SOURCE" redefined
In file included from 
/usr/local/lib/gcc-lib/alphaev67-dec-osf5.1/3.3.3/include/string.h:61,
                  from Lib/cluster/src/vq_wrap.cpp:27:
/usr/include/standards.h:198:1: warning: this is the location of the 
previous definition
In file included from 
/home2/menut/prg/python2.3/include/python2.3/Python.h:8,
                  from Lib/cluster/src/vq_wrap.cpp:176:
/home2/menut/prg/python2.3/include/python2.3/pyconfig.h:859:1: warning: 
"_XOPEN_SOURCE" redefined
In file included from 
/usr/local/lib/gcc-lib/alphaev67-dec-osf5.1/3.3.3/include/string.h:61,
                  from Lib/cluster/src/vq_wrap.cpp:27:
/usr/include/standards.h:188:1: warning: this is the location of the 
previous definition
In file included from 
/home2/menut/prg/python2.3/include/python2.3/pyport.h:99,
                  from 
/home2/menut/prg/python2.3/include/python2.3/Python.h:48,
                  from Lib/cluster/src/vq_wrap.cpp:176:
/usr/include/sys/time.h:71: error: 'suseconds_t' is used as a type, but 
is not
    defined as a type.
In file included from 
/home2/menut/prg/python2.3/include/python2.3/pyport.h:157,
                  from 
/home2/menut/prg/python2.3/include/python2.3/Python.h:48,
                  from Lib/cluster/src/vq_wrap.cpp:176:
/usr/local/lib/gcc-lib/alphaev67-dec-osf5.1/3.3.3/include/sys/stat.h:213: 
error: '
    blksize_t' is used as a type, but is not defined as a type.
/usr/local/lib/gcc-lib/alphaev67-dec-osf5.1/3.3.3/include/sys/stat.h:213: 
error: '
    blkcnt_t' is used as a type, but is not defined as a type.
In file included from Lib/cluster/src/vq_wrap.cpp:499:
Lib/cluster/src/vq.h:57:7: warning: no newline at end of file
In file included from 
/home2/menut/prg/python2.3/include/python2.3/Python.h:8,
                  from Lib/cluster/src/vq_wrap.cpp:176:
/home2/menut/prg/python2.3/include/python2.3/pyconfig.h:840:1: warning: 
"_OSF_SOURCE" redefined
In file included from 
/usr/local/lib/gcc-lib/alphaev67-dec-osf5.1/3.3.3/include/string.h:61,
                  from Lib/cluster/src/vq_wrap.cpp:27:
/usr/include/standards.h:206:1: warning: this is the location of the 
previous definition
In file included from 
/home2/menut/prg/python2.3/include/python2.3/Python.h:8,
                  from Lib/cluster/src/vq_wrap.cpp:176:
/home2/menut/prg/python2.3/include/python2.3/pyconfig.h:847:1: warning: 
"_POSIX_C_SOURCE" redefined
In file included from 
/usr/local/lib/gcc-lib/alphaev67-dec-osf5.1/3.3.3/include/string.h:61,
                  from Lib/cluster/src/vq_wrap.cpp:27:
/usr/include/standards.h:198:1: warning: this is the location of the 
previous definition
In file included from 
/home2/menut/prg/python2.3/include/python2.3/Python.h:8,
                  from Lib/cluster/src/vq_wrap.cpp:176:
/home2/menut/prg/python2.3/include/python2.3/pyconfig.h:859:1: warning: 
"_XOPEN_SOURCE" redefined
In file included from 
/usr/local/lib/gcc-lib/alphaev67-dec-osf5.1/3.3.3/include/string.h:61,
                  from Lib/cluster/src/vq_wrap.cpp:27:
/usr/include/standards.h:188:1: warning: this is the location of the 
previous definition
In file included from 
/home2/menut/prg/python2.3/include/python2.3/pyport.h:99,
                  from 
/home2/menut/prg/python2.3/include/python2.3/Python.h:48,
                  from Lib/cluster/src/vq_wrap.cpp:176:
/usr/include/sys/time.h:71: error: 'suseconds_t' is used as a type, but 
is not
    defined as a type.
In file included from 
/home2/menut/prg/python2.3/include/python2.3/pyport.h:157,
                  from 
/home2/menut/prg/python2.3/include/python2.3/Python.h:48,
                  from Lib/cluster/src/vq_wrap.cpp:176:
/usr/local/lib/gcc-lib/alphaev67-dec-osf5.1/3.3.3/include/sys/stat.h:213: 
error: '
    blksize_t' is used as a type, but is not defined as a type.
/usr/local/lib/gcc-lib/alphaev67-dec-osf5.1/3.3.3/include/sys/stat.h:213: 
error: '
    blkcnt_t' is used as a type, but is not defined as a type.
In file included from Lib/cluster/src/vq_wrap.cpp:499:
Lib/cluster/src/vq.h:57:7: warning: no newline at end of file
error: Command "c++ -pthread -fno-strict-aliasing -DNDEBUG -g -O3 -Wall 
-Wstrict-prototypes -I/home2/menut/prg/python2.3/include/python2.3 -c 
Lib/cluster/src/vq_wrap.cpp -o 
build/temp.osf1-V5.1-alpha-2.3/Lib/cluster/src/vq_wrap.o" failed with 
exit status 1



From oliphant at ee.byu.edu  Fri May 21 15:40:13 2004
From: oliphant at ee.byu.edu (Travis E. Oliphant)
Date: Fri, 21 May 2004 13:40:13 -0600
Subject: [SciPy-user] distribution.fit() methods
In-Reply-To: <40AE0564.10706@psychology.nottingham.ac.uk>
References: <mailman.0.1085147088.18550.scipy-user@scipy.net>
	<40AE0564.10706@psychology.nottingham.ac.uk>
Message-ID: <40AE5B1D.5060501@ee.byu.edu>

Jon Peirce wrote:
> Hi there,
> 
> I'm trying to fit a weibull function to some data using Scipy. It looks 
> like I should be able to use exponweibull_gen.fit but I'm a bit puzzled 
> about the arguments. According to the code the arguments  (data, *args, 
> **kwds) are needed, where *args seems to contain the shape parameters 
> for the fit and **kwds contains the location and scale params.
> 
> But if I'm fitting a function, surely I want to give xData and yData and 
> *not* shape arguments (or maybe a first guess at those). Does anyone 
> know how to use these methods?
> 

These fit methods are a fairly thin wrapper around optimize.fmin.  The 
intended use is to take random variates you suspect as being weibull 
distributed and find the maximum likelihood estimates of the shape, 
location, and scale parameters.

Notice, it does not try to fit (x,y) coordinates to a weibull function.
To do that you should just use an optimization method and with a fit 
function that involves the exponweib.pdf function.

Please clarify what it is that you are trying to do.   Also, the 
XXXXX_gen classes are not intended for general use unless you know what 
you are doing.  What you want is exponweib.XXX --- this gives you a 
realization of the XXXXX_gen class.

-Travis O.



From jonathan.peirce at nottingham.ac.uk  Sat May 22 09:33:02 2004
From: jonathan.peirce at nottingham.ac.uk (Jon Peirce)
Date: Sat, 22 May 2004 14:33:02 +0100
Subject: [SciPy-user] Re: distribution.fit() methods
In-Reply-To: <40AE5B1D.5060501@ee.byu.edu>
References: <mailman.0.1085147088.18550.scipy-user@scipy.net>
	<40AE0564.10706@psychology.nottingham.ac.uk>
	<40AE5B1D.5060501@ee.byu.edu>
Message-ID: <c8nkq8$gid$1@sea.gmane.org>

Travis E. Oliphant wrote:
> Jon Peirce wrote:
> 
>> Hi there,
>>
>> I'm trying to fit a weibull function to some data using Scipy. It 
>> looks like I should be able to use exponweibull_gen.fit but I'm a bit 
>> puzzled about the arguments. According to the code the arguments  
>> (data, *args, **kwds) are needed, where *args seems to contain the 
>> shape parameters for the fit and **kwds contains the location and 
>> scale params.
>>
>> But if I'm fitting a function, surely I want to give xData and yData 
>> and *not* shape arguments (or maybe a first guess at those). Does 
>> anyone know how to use these methods?
>>
> 
> These fit methods are a fairly thin wrapper around optimize.fmin.  The 
> intended use is to take random variates you suspect as being weibull 
> distributed and find the maximum likelihood estimates of the shape, 
> location, and scale parameters.
> 
> Notice, it does not try to fit (x,y) coordinates to a weibull function.
> To do that you should just use an optimization method and with a fit 
> function that involves the exponweib.pdf function.
> 
> Please clarify what it is that you are trying to do.   Also, the 
> XXXXX_gen classes are not intended for general use unless you know what 
> you are doing.  What you want is exponweib.XXX --- this gives you a 
> realization of the XXXXX_gen class.
> 
> -Travis O.

Thanks (and sorry my post was kinda garbled - head wasn't screwed on 
tight enough yesterday!!)

I am now using exponweib.pdf and am able to use optimize.fmin to get the 
appropriate shape parameters - that's what I thought xxx.fit was doing, 
but I see now that it expects a raw variate rather than the pdf or cdf 
coordinates.

chrs, Jon



From nwagner at mecha.uni-stuttgart.de  Mon May 24 04:29:20 2004
From: nwagner at mecha.uni-stuttgart.de (Nils Wagner)
Date: Mon, 24 May 2004 10:29:20 +0200
Subject: [SciPy-user] dpi size of an exisiting graphics window
Message-ID: <40B1B260.9050800@mecha.uni-stuttgart.de>

Dear experts,

Is it possible to change the dpi size of an existing graphics window ?

Nils



From ariciputi at pito.com  Tue May 25 06:40:04 2004
From: ariciputi at pito.com (Andrea Riciputi)
Date: Tue, 25 May 2004 12:40:04 +0200
Subject: [SciPy-user] Data convolution.
Message-ID: <E237E707-AE37-11D8-9C01-000A95C0BC68@pito.com>

Hi,
I need to perform convolution on a set of two dimensional data 
resulting from a numerical simulation. Since my numerical simulation is 
written in Python and C (for CPU intensive task), I'm wondering if 
scipy is able to fit my needs. I've read the tutorial but it's quite 
terse about the topic, further I've found a reference to FFT section 
that has not been written.

Could anyone help me with some hints?

Thanks in advance,
	Andrea.



From nwagner at mecha.uni-stuttgart.de  Wed May 26 06:58:11 2004
From: nwagner at mecha.uni-stuttgart.de (Nils Wagner)
Date: Wed, 26 May 2004 12:58:11 +0200
Subject: [SciPy-user] Schur complement
Message-ID: <40B47843.2000005@mecha.uni-stuttgart.de>

Dear experts,

I am trying to solve a transcendental eigenvalue problem via the Schur 
complement. However, I have some trouble with optimize.fsolve.

How can I solve this problem ?

Thanks in advance.

Nils

Traceback (most recent call last):
   File "fschur.py", line 58, in ?
     S = optimize.fsolve(func,x0)
   File "/usr/lib/python2.3/site-packages/scipy/optimize/minpack.py", 
line 80, in fsolve
     check_func(func,x0,args,n,(n,))
   File 
"/usr/lib/python2.3/site-packages/scipy/optimize/common_routines.py", 
line 13, in check_func
     res = myasarray(apply(thefunc,args))
   File "fschur.py", line 56, in func
     return schur(T(x))
   File "fschur.py", line 8, in T
     tmp[0,0] = -sin(x)
ValueError: array too large for destination
 >>>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: fschur.py
Type: text/x-python
Size: 967 bytes
Desc: not available
URL: <http://mail.scipy.org/pipermail/scipy-user/attachments/20040526/c9c9b281/attachment.py>

From nwagner at mecha.uni-stuttgart.de  Wed May 26 08:00:58 2004
From: nwagner at mecha.uni-stuttgart.de (Nils Wagner)
Date: Wed, 26 May 2004 14:00:58 +0200
Subject: [SciPy-user] brentq, brenth in case of complex roots 
Message-ID: <40B486FA.6060705@mecha.uni-stuttgart.de>

Dear experts,

How can I apply brentq, brenth etc. in case of complex functions ?

Nils



From pearu at scipy.org  Wed May 26 10:46:07 2004
From: pearu at scipy.org (Pearu Peterson)
Date: Wed, 26 May 2004 09:46:07 -0500 (CDT)
Subject: [SciPy-user] Schur complement
In-Reply-To: <40B47843.2000005@mecha.uni-stuttgart.de>
References: <40B47843.2000005@mecha.uni-stuttgart.de>
Message-ID: <Pine.LNX.4.58.0405260943170.16271@scipy.org>



On Wed, 26 May 2004, Nils Wagner wrote:

> I am trying to solve a transcendental eigenvalue problem via the Schur 
> complement. However, I have some trouble with optimize.fsolve.
> 
> How can I solve this problem ?

In short, use

def func(x):
   return abs(schur(T(x[0])))

(when using optimize.fsolve then the argument x is rank-1 array and
fsolve assumes that func is a real-valued function)

Pearu



From nwagner at mecha.uni-stuttgart.de  Wed May 26 10:52:20 2004
From: nwagner at mecha.uni-stuttgart.de (Nils Wagner)
Date: Wed, 26 May 2004 16:52:20 +0200
Subject: [SciPy-user] Schur complement
In-Reply-To: <Pine.LNX.4.58.0405260943170.16271@scipy.org>
References: <40B47843.2000005@mecha.uni-stuttgart.de>
	<Pine.LNX.4.58.0405260943170.16271@scipy.org>
Message-ID: <40B4AF24.7020603@mecha.uni-stuttgart.de>

Pearu Peterson wrote:
> 
> On Wed, 26 May 2004, Nils Wagner wrote:
> 
> 
>>I am trying to solve a transcendental eigenvalue problem via the Schur 
>>complement. However, I have some trouble with optimize.fsolve.
>>
>>How can I solve this problem ?
> 
> 
> In short, use
> 
> def func(x):
>    return abs(schur(T(x[0])))
> 
> (when using optimize.fsolve then the argument x is rank-1 array and
> fsolve assumes that func is a real-valued function)
> 
> Pearu
> 
> _______________________________________________
> SciPy-user mailing list
> SciPy-user at scipy.net
> http://www.scipy.net/mailman/listinfo/scipy-user
> 
I have modified my program.

def func2(x):
  return abs(schur(T(x)))

r6 = optimize.fsolve(func2,x0)

Now I get

Traceback (most recent call last):
   File "fschur.py", line 77, in ?
     r6 = optimize.fsolve(func2,x0)
   File "/usr/lib/python2.3/site-packages/scipy/optimize/minpack.py", 
line 80, in fsolve
     check_func(func,x0,args,n,(n,))
   File 
"/usr/lib/python2.3/site-packages/scipy/optimize/common_routines.py", 
line 13, in check_func
     res = myasarray(apply(thefunc,args))
   File "fschur.py", line 63, in func2
     return abs(schur(T(x)))
   File "fschur.py", line 8, in T
     tmp[0,0] = -sin(x)
ValueError: array too large for destination
 >>>


The requirement, i.e. func is real is not mentioned in help 
(optimize.fsolve). Am I missing something ?


Nils

fsolve(func, x0, args=(), fprime=None, full_output=0, col_deriv=0, 
xtol=1.49012e
-08, maxfev=0, band=None, epsfcn=0.0, factor=100, diag=None)
     Find the roots of a function.

     Description:

       Return the roots of the (non-linear) equations defined by
       func(x)=0 given a starting estimate.

     Inputs:

       func -- A Python function or method which takes at least one
               (possibly vector) argument.
       x0 -- The starting estimate for the roots of func(x)=0.
       args -- Any extra arguments to func are placed in this tuple.
       fprime -- A function or method to compute the Jacobian of func with
               derivatives across the rows. If this is None, the
               Jacobian will be estimated.
       full_output -- non-zero to return the optional outputs.
       col_deriv -- non-zero to specify that the Jacobian function
                    computes derivatives down the columns (faster, because
                    there is no transpose operation).

     Outputs: (x, {infodict, ier, mesg})

       x -- the solution (or the result of the last iteration for an
            unsuccessful call.

       infodict -- a dictionary of optional outputs with the keys:
                   'nfev' : the number of function calls
                   'njev' : the number of jacobian calls
                   'fvec' : the function evaluated at the output
                   'fjac' : the orthogonal matrix, q, produced by the
                            QR facotrization of the final approximate
                            Jacobi matrix, stored column wise.
                   'r'    : upper triangular matrix produced by QR
                            factorization of same matrix.
                   'qtf'  : the vector (transpose(q) * fvec).
       ier -- an integer flag.  If it is equal to 1 the solution was
              found.  If it is not equal to 1, the solution was not
              found and the following message gives more information.
       mesg -- a string message giving information about the cause of
               failure.

     Extended Inputs:

      xtol -- The calculation will terminate if the relative error
              between two consecutive iterates is at most xtol.
      maxfev -- The maximum number of calls to the function. If zero,
                then 100*(N+1) is the maximum where N is the number
                of elements in x0.
      band -- If set to a two-sequence containing the number of sub-
              and superdiagonals within the band of the Jacobi matrix,




From pearu at scipy.org  Wed May 26 11:14:22 2004
From: pearu at scipy.org (Pearu Peterson)
Date: Wed, 26 May 2004 10:14:22 -0500 (CDT)
Subject: [SciPy-user] Schur complement
In-Reply-To: <40B4AF24.7020603@mecha.uni-stuttgart.de>
References: <40B47843.2000005@mecha.uni-stuttgart.de>
	<40B4AF24.7020603@mecha.uni-stuttgart.de>
Message-ID: <Pine.LNX.4.58.0405261007171.16271@scipy.org>



On Wed, 26 May 2004, Nils Wagner wrote:

> > In short, use
> > 
> > def func(x):
> >    return abs(schur(T(x[0])))
                           ^^^

> > 
> I have modified my program.
> 
> def func2(x):
>   return abs(schur(T(x)))
> 
> r6 = optimize.fsolve(func2,x0)
> 
> Now I get
> 
> Traceback (most recent call last):
> ValueError: array too large for destination

Compare my suggestion with your version of func2 (notice ^^^).

> The requirement, i.e. func is real is not mentioned in help 
> (optimize.fsolve). Am I missing something ?

Notice that it also does not mention that func can be complex valued.
Anyways, fsolve assumes real-valued functions - `.`. But it is also simple 
to use fsolve for complex-valued functions: make sure that func returns real 
and imaginary parts of a function value, also an initial value to 
fsolve should be specified as a pair of real and imaginary parts of the 
value.

Pearu



From nwagner at mecha.uni-stuttgart.de  Wed May 26 11:30:11 2004
From: nwagner at mecha.uni-stuttgart.de (Nils Wagner)
Date: Wed, 26 May 2004 17:30:11 +0200
Subject: [SciPy-user] Schur complement
In-Reply-To: <Pine.LNX.4.58.0405261007171.16271@scipy.org>
References: <40B47843.2000005@mecha.uni-stuttgart.de>
	<40B4AF24.7020603@mecha.uni-stuttgart.de>
	<Pine.LNX.4.58.0405261007171.16271@scipy.org>
Message-ID: <40B4B803.8090907@mecha.uni-stuttgart.de>

Pearu Peterson wrote:
> 
> On Wed, 26 May 2004, Nils Wagner wrote:
> 
> 
>>>In short, use
>>>
>>>def func(x):
>>>   return abs(schur(T(x[0])))
> 
>                            ^^^
> 
> 
>>I have modified my program.
>>
>>def func2(x):
>>  return abs(schur(T(x)))
>>
>>r6 = optimize.fsolve(func2,x0)
>>
>>Now I get
>>
>>Traceback (most recent call last):
>>ValueError: array too large for destination
> 
> 
> Compare my suggestion with your version of func2 (notice ^^^).
> 
> 
>>The requirement, i.e. func is real is not mentioned in help 
>>(optimize.fsolve). Am I missing something ?
> 
> 
> Notice that it also does not mention that func can be complex valued.
> Anyways, fsolve assumes real-valued functions - `.`. But it is also simple 
> to use fsolve for complex-valued functions: make sure that func returns real 
> and imaginary parts of a function value, also an initial value to 
> fsolve should be specified as a pair of real and imaginary parts of the 
> value.
> 
Dear Pearu,

Thank you very much for your help. Moreover, it would be very kind of 
you, if you could send me a short example for complex-valued functions 
as well.

Thanks in advance.

       Nils

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



From pearu at scipy.org  Wed May 26 11:49:23 2004
From: pearu at scipy.org (Pearu Peterson)
Date: Wed, 26 May 2004 10:49:23 -0500 (CDT)
Subject: [SciPy-user] Schur complement
In-Reply-To: <40B4B803.8090907@mecha.uni-stuttgart.de>
References: <40B47843.2000005@mecha.uni-stuttgart.de>
	<40B4AF24.7020603@mecha.uni-stuttgart.de>
	<40B4B803.8090907@mecha.uni-stuttgart.de>
Message-ID: <Pine.LNX.4.58.0405261048350.16271@scipy.org>



On Wed, 26 May 2004, Nils Wagner wrote:

> Thank you very much for your help. Moreover, it would be very kind of 
> you, if you could send me a short example for complex-valued functions 
> as well.

In [1]: from scipy import *

In [2]: def cfunc(z): return z+2j
   ...: 

In [3]: def rfunc2(x): 
   ...:   v = cfunc(x[0]+x[1]*1j)
   ...:   return v.real,v.imag
   ...: 

In [4]: optimize.fsolve(rfunc2,(3,1))
Out[4]: array([ -8.07793567e-28,  -2.00000000e+00])

Pearu



From nwagner at mecha.uni-stuttgart.de  Wed May 26 11:53:56 2004
From: nwagner at mecha.uni-stuttgart.de (Nils Wagner)
Date: Wed, 26 May 2004 17:53:56 +0200
Subject: [SciPy-user] Schur complement
In-Reply-To: <Pine.LNX.4.58.0405261048350.16271@scipy.org>
References: <40B47843.2000005@mecha.uni-stuttgart.de>
	<40B4AF24.7020603@mecha.uni-stuttgart.de>
	<40B4B803.8090907@mecha.uni-stuttgart.de>
	<Pine.LNX.4.58.0405261048350.16271@scipy.org>
Message-ID: <40B4BD94.1090207@mecha.uni-stuttgart.de>

Pearu Peterson wrote:
> 
> On Wed, 26 May 2004, Nils Wagner wrote:
> 
> 
>>Thank you very much for your help. Moreover, it would be very kind of 
>>you, if you could send me a short example for complex-valued functions 
>>as well.
> 
> 
> In [1]: from scipy import *
> 
> In [2]: def cfunc(z): return z+2j
>    ...: 
> 
> In [3]: def rfunc2(x): 
>    ...:   v = cfunc(x[0]+x[1]*1j)
>    ...:   return v.real,v.imag
>    ...: 
> 
> In [4]: optimize.fsolve(rfunc2,(3,1))
> Out[4]: array([ -8.07793567e-28,  -2.00000000e+00])
> 
> Pearu
> 
> _______________________________________________
> SciPy-user mailing list
> SciPy-user at scipy.net
> http://www.scipy.net/mailman/listinfo/scipy-user
> 
Hopefully my last question with respect to this topic ...
Can I apply brentq, brenth in case of complex-valued functions ?

Nils



From pearu at scipy.org  Wed May 26 13:01:44 2004
From: pearu at scipy.org (Pearu Peterson)
Date: Wed, 26 May 2004 12:01:44 -0500 (CDT)
Subject: [SciPy-user] Schur complement
In-Reply-To: <40B4BD94.1090207@mecha.uni-stuttgart.de>
References: <40B47843.2000005@mecha.uni-stuttgart.de>
	<40B4AF24.7020603@mecha.uni-stuttgart.de>
	<Pine.LNX.4.58.0405261048350.16271@scipy.org>
	<40B4BD94.1090207@mecha.uni-stuttgart.de>
Message-ID: <Pine.LNX.4.58.0405261157440.16271@scipy.org>



On Wed, 26 May 2004, Nils Wagner wrote:

> Hopefully my last question with respect to this topic ...
> Can I apply brentq, brenth in case of complex-valued functions ?

No. These functions can only deal with real-valued functions with one 
real argument.

Pearu



From fonnesbeck at mac.com  Wed May 26 12:56:31 2004
From: fonnesbeck at mac.com (Christopher Fonnesbeck)
Date: Wed, 26 May 2004 12:56:31 -0400
Subject: [SciPy-user] build errors on OSX using IBM XL FORTRAN
Message-ID: <A304E67C-AF35-11D8-AD48-000A956FDAC0@mac.com>

Using today's cvs update, I get the following error when building on  
OSX 10.3; seems to be related to XL FORTRAN:

compling Fortran sources
xlf(f77) options: '-qextname -O5'
xlf90(f90) options: '-qextname -O5'
xlf90(fix) options: '-qfixed -qextname -O5'
creating build/temp.darwin-7.3.0-Power_Macintosh-2.3/Lib/linalg
creating build/temp.darwin-7.3.0-Power_Macintosh-2.3/Lib/linalg/src
compile options: '-DNO_ATLAS_INFO=3 -Ibuild/src  
-I/System/Library/Frameworks/Python.framework/Versions/2.3/include/ 
python2.3 -c'
xlf:f77: Lib/linalg/src/fblaswrap.f
xlf: 1501-210 command option NO_ATLAS_INFO=3 contains an incorrect  
subargument
xlf: 1501-210 command option NO_ATLAS_INFO=3 contains an incorrect  
subargument
error: Command "xlf -qextname -O5 -DNO_ATLAS_INFO=3 -Ibuild/src  
-I/System/Library/Frameworks/Python.framework/Versions/2.3/include/ 
python2.3 -c -c Lib/linalg/src/fblaswrap.f -o  
build/temp.darwin-7.3.0-Power_Macintosh-2.3/Lib/linalg/src/fblaswrap.o"  
failed with exit status 40


--
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

"I don't remember any kind of heaviness ruining my time at Yale."
          George W. Bush, on the civil rights movement
--
Putting http://wecanstopspam.org in your email helps it pass through  
overzealous spam filters. 



From pearu at scipy.org  Wed May 26 13:32:14 2004
From: pearu at scipy.org (Pearu Peterson)
Date: Wed, 26 May 2004 12:32:14 -0500 (CDT)
Subject: [SciPy-user] build errors on OSX using IBM XL FORTRAN
In-Reply-To: <A304E67C-AF35-11D8-AD48-000A956FDAC0@mac.com>
References: <A304E67C-AF35-11D8-AD48-000A956FDAC0@mac.com>
Message-ID: <Pine.LNX.4.58.0405261230230.16271@scipy.org>



On Wed, 26 May 2004, Christopher Fonnesbeck wrote:

> Using today's cvs update, I get the following error when building on  
> OSX 10.3; seems to be related to XL FORTRAN:
> 
> compling Fortran sources
> xlf(f77) options: '-qextname -O5'
> xlf90(f90) options: '-qextname -O5'
> xlf90(fix) options: '-qfixed -qextname -O5'
> creating build/temp.darwin-7.3.0-Power_Macintosh-2.3/Lib/linalg
> creating build/temp.darwin-7.3.0-Power_Macintosh-2.3/Lib/linalg/src
> compile options: '-DNO_ATLAS_INFO=3 -Ibuild/src  
> -I/System/Library/Frameworks/Python.framework/Versions/2.3/include/ 
> python2.3 -c'
> xlf:f77: Lib/linalg/src/fblaswrap.f
> xlf: 1501-210 command option NO_ATLAS_INFO=3 contains an incorrect  
> subargument
> xlf: 1501-210 command option NO_ATLAS_INFO=3 contains an incorrect  
> subargument
> error: Command "xlf -qextname -O5 -DNO_ATLAS_INFO=3 -Ibuild/src  
> -I/System/Library/Frameworks/Python.framework/Versions/2.3/include/ 
> python2.3 -c -c Lib/linalg/src/fblaswrap.f -o  
> build/temp.darwin-7.3.0-Power_Macintosh-2.3/Lib/linalg/src/fblaswrap.o"  
> failed with exit status 40

Try again. Hopefully the patch in CVS fixes this problem. It turns out 
that -D/-U options to xlf compiler are not for defining cpp macros...

Pearu



From fonnesbeck at mac.com  Wed May 26 14:06:53 2004
From: fonnesbeck at mac.com (Christopher Fonnesbeck)
Date: Wed, 26 May 2004 14:06:53 -0400
Subject: [SciPy-user] build errors on OSX using IBM XL FORTRAN
In-Reply-To: <Pine.LNX.4.58.0405261230230.16271@scipy.org>
References: <A304E67C-AF35-11D8-AD48-000A956FDAC0@mac.com>
	<Pine.LNX.4.58.0405261230230.16271@scipy.org>
Message-ID: <77DE7F33-AF3F-11D8-AD48-000A956FDAC0@mac.com>


On May 26, 2004, at 1:32 PM, Pearu Peterson wrote:

>
> Try again. Hopefully the patch in CVS fixes this problem. It turns out
> that -D/-U options to xlf compiler are not for defining cpp macros...
>
> Pearu
>

This appears to have worked ... 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

"I don't remember any kind of heaviness ruining my time at Yale."
          George W. Bush, on the civil rights movement
--
Putting http://wecanstopspam.org in your email helps it pass through 
overzealous spam filters.



From oliphant at ee.byu.edu  Wed May 26 14:33:33 2004
From: oliphant at ee.byu.edu (Travis E. Oliphant)
Date: Wed, 26 May 2004 12:33:33 -0600
Subject: [SciPy-user] Data convolution.
In-Reply-To: <E237E707-AE37-11D8-9C01-000A95C0BC68@pito.com>
References: <E237E707-AE37-11D8-9C01-000A95C0BC68@pito.com>
Message-ID: <40B4E2FD.9040104@ee.byu.edu>

Andrea Riciputi wrote:
> Hi,
> I need to perform convolution on a set of two dimensional data resulting 
> from a numerical simulation. Since my numerical simulation is written in 
> Python and C (for CPU intensive task), I'm wondering if scipy is able to 
> fit my needs. I've read the tutorial but it's quite terse about the 
> topic, further I've found a reference to FFT section that has not been 
> written.
> 

Yes, you can do convolution.

from scipy import *

info(signal.convolve)

should give more information.  But... this is really for convolution 
where one or both of the inputs are small for 2.D. FIR filtering (say 
less than 32x32 -- not really sure of the cutoff size).

Convolution of large data sets can be done  be done with FFT's (but 
their will be boundary wrapping issues that you will need to address --- 
or ignore depending on what you are doing).

To get something to compare with

c = signal.convolve(a,b)

will do the job.  An optional third input will control the size of the 
resulting output.  Post if you have further questions.

-Travis O.





From ariciputi at pito.com  Wed May 26 18:18:30 2004
From: ariciputi at pito.com (Andrea Riciputi)
Date: Thu, 27 May 2004 00:18:30 +0200
Subject: [SciPy-user] Data convolution.
In-Reply-To: <40B4E2FD.9040104@ee.byu.edu>
References: <E237E707-AE37-11D8-9C01-000A95C0BC68@pito.com>
	<40B4E2FD.9040104@ee.byu.edu>
Message-ID: <9E3C17E8-AF62-11D8-BF3D-000A95C0BC68@pito.com>

On 26 May 2004, at 20:33, Travis E. Oliphant wrote:

> Yes, you can do convolution.
>
> from scipy import *
>
> info(signal.convolve)
>
> [snip]

Thanks, but my data set are typically 1024x1024 so I think I need FFT. 
Nevertheless I'll look into scipy for testing and verifying my own 
code.

Cheers,
	Andrea.



From oliphant at ee.byu.edu  Wed May 26 19:01:24 2004
From: oliphant at ee.byu.edu (Travis E. Oliphant)
Date: Wed, 26 May 2004 17:01:24 -0600
Subject: [SciPy-user] Data convolution.
In-Reply-To: <9E3C17E8-AF62-11D8-BF3D-000A95C0BC68@pito.com>
References: <E237E707-AE37-11D8-9C01-000A95C0BC68@pito.com>
	<40B4E2FD.9040104@ee.byu.edu>
	<9E3C17E8-AF62-11D8-BF3D-000A95C0BC68@pito.com>
Message-ID: <40B521C4.6070306@ee.byu.edu>

Andrea Riciputi wrote:
> On 26 May 2004, at 20:33, Travis E. Oliphant wrote:
> 
>> Yes, you can do convolution.
>>
>> from scipy import *
>>
>> info(signal.convolve)
>>
>> [snip]
> 
> 
> Thanks, but my data set are typically 1024x1024 so I think I need FFT. 
> Nevertheless I'll look into scipy for testing and verifying my own code.
> 

info(fft)

will give you loads of information on how to use fft's to do convolution.

It could be as simple as

ifft2(fft2(a)*fft2(b))

depending on your boundary issues.

-Travis O.






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



From nwagner at mecha.uni-stuttgart.de  Thu May 27 03:49:29 2004
From: nwagner at mecha.uni-stuttgart.de (Nils Wagner)
Date: Thu, 27 May 2004 09:49:29 +0200
Subject: [SciPy-user] Rootfinding procedures for Vector-valued functions 
In-Reply-To: <Pine.LNX.4.58.0405261157440.16271@scipy.org>
References: <40B47843.2000005@mecha.uni-stuttgart.de>
	<40B4AF24.7020603@mecha.uni-stuttgart.de>
	<Pine.LNX.4.58.0405261048350.16271@scipy.org>
	<40B4BD94.1090207@mecha.uni-stuttgart.de>
	<Pine.LNX.4.58.0405261157440.16271@scipy.org>
Message-ID: <40B59D89.8000303@mecha.uni-stuttgart.de>

Pearu Peterson wrote:
> 
> On Wed, 26 May 2004, Nils Wagner wrote:
> 
> 
>>Hopefully my last question with respect to this topic ...
>>Can I apply brentq, brenth in case of complex-valued functions ?
> 
> 
> No. These functions can only deal with real-valued functions with one 
> real argument.
> 
> Pearu
> 
> _______________________________________________
> SciPy-user mailing list
> SciPy-user at scipy.net
> http://www.scipy.net/mailman/listinfo/scipy-user
> 

Is it planned to upgrade the functions with respect to complex-valued 
functions and vector-valued functions respectively ?

Nils




From stephen.walton at csun.edu  Thu May 27 17:45:45 2004
From: stephen.walton at csun.edu (Stephen Walton)
Date: Thu, 27 May 2004 14:45:45 -0700
Subject: [SciPy-user] Data convolution.
In-Reply-To: <40B521C4.6070306@ee.byu.edu>
References: <E237E707-AE37-11D8-9C01-000A95C0BC68@pito.com>
	<40B4E2FD.9040104@ee.byu.edu>
	<9E3C17E8-AF62-11D8-BF3D-000A95C0BC68@pito.com>
	<40B521C4.6070306@ee.byu.edu>
Message-ID: <1085694345.6683.12.camel@s123n051.sfo.csun.edu>

On Wed, 2004-05-26 at 16:01, Travis E. Oliphant wrote:
> Andrea Riciputi wrote:
> > 
> > Thanks, but my data set are typically 1024x1024 so I think I need FFT. 
> > Nevertheless I'll look into scipy for testing and verifying my own code.
> > 

> It could be as simple as
> 
> ifft2(fft2(a)*fft2(b))

I've got quite a bit of experience doing this with astronomical images. 
Travis is right about boundary issues;  you almost certainly want to
extend your images to twice their original size in each direction (four
times their original area) with the mean of your images.  In my case,
when I do this with full disk images of the Sun, I extend the image with
a model aureole to avoid edge effects.  This may be overkill.

All my code for doing this is in Fortran and built on top of the PORT
library FFT routines, so you probably don't want a copy.

-- 
Stephen Walton <stephen.walton at csun.edu>
Dept. of Physics & Astronomy, Cal State Northridge



From cookedm at physics.mcmaster.ca  Thu May 27 23:24:34 2004
From: cookedm at physics.mcmaster.ca (David M. Cooke)
Date: Thu, 27 May 2004 23:24:34 -0400
Subject: [SciPy-user] Rootfinding procedures for Vector-valued functions
In-Reply-To: <40B59D89.8000303@mecha.uni-stuttgart.de>
References: <40B47843.2000005@mecha.uni-stuttgart.de>
	<40B4AF24.7020603@mecha.uni-stuttgart.de>
	<Pine.LNX.4.58.0405261048350.16271@scipy.org>
	<40B4BD94.1090207@mecha.uni-stuttgart.de>
	<Pine.LNX.4.58.0405261157440.16271@scipy.org>
	<40B59D89.8000303@mecha.uni-stuttgart.de>
Message-ID: <20040528032434.GA21308@arbutus.physics.mcmaster.ca>

On Thu, May 27, 2004 at 09:49:29AM +0200, Nils Wagner wrote:
> Pearu Peterson wrote:
> >
> >On Wed, 26 May 2004, Nils Wagner wrote:
> >
> >
> >>Hopefully my last question with respect to this topic ...
> >>Can I apply brentq, brenth in case of complex-valued functions ?
> >
> >
> >No. These functions can only deal with real-valued functions with one 
> >real argument.
> >
> >Pearu
> 
> Is it planned to upgrade the functions with respect to complex-valued 
> functions and vector-valued functions respectively ?

How? The Brent routines are one-dimensional zero finders.
Multidimensional (which includes complex) zero finders are much
different beasts from one-dimensional routines -- in general, you don't
a bracket around the zero.

You want scipy.optimize.fsolve for multidimensional problems.

-- 
|>|\/|<
/--------------------------------------------------------------------------\
|David M. Cooke                      http://arbutus.physics.mcmaster.ca/dmc/
|cookedm at physics.mcmaster.ca



From lanceboyle at cwazy.co.uk  Fri May 28 00:46:41 2004
From: lanceboyle at cwazy.co.uk (Lance Boyle)
Date: Thu, 27 May 2004 21:46:41 -0700
Subject: [SciPy-user] Data convolution.
In-Reply-To: <1085694345.6683.12.camel@s123n051.sfo.csun.edu>
References: <E237E707-AE37-11D8-9C01-000A95C0BC68@pito.com>
	<40B4E2FD.9040104@ee.byu.edu>
	<9E3C17E8-AF62-11D8-BF3D-000A95C0BC68@pito.com>
	<40B521C4.6070306@ee.byu.edu>
	<1085694345.6683.12.camel@s123n051.sfo.csun.edu>
Message-ID: <031E40B9-B062-11D8-B39D-003065F93FF0@cwazy.co.uk>

When doing (linear) convolutions (of any dimension) with FFTs, does 
SciPy take care of aliasing in the "other" domain, usually time domain 
and spatial domain? This is done by zero-padding and with the so-called 
overlap-add and overlap-save methods. This is a different issue from 
edge effects (as I understand the earlier posters) which are Gibbs 
phenomena.

Jerry


On May 27, 2004, at 2:45 PM, Stephen Walton wrote:

> On Wed, 2004-05-26 at 16:01, Travis E. Oliphant wrote:
>> Andrea Riciputi wrote:
>>>
>>> Thanks, but my data set are typically 1024x1024 so I think I need 
>>> FFT.
>>> Nevertheless I'll look into scipy for testing and verifying my own 
>>> code.
>>>
>
>> It could be as simple as
>>
>> ifft2(fft2(a)*fft2(b))
>
> I've got quite a bit of experience doing this with astronomical images.
> Travis is right about boundary issues;  you almost certainly want to
> extend your images to twice their original size in each direction (four
> times their original area) with the mean of your images.  In my case,
> when I do this with full disk images of the Sun, I extend the image 
> with
> a model aureole to avoid edge effects.  This may be overkill.
>
> All my code for doing this is in Fortran and built on top of the PORT
> library FFT routines, so you probably don't want a copy.
>
> -- 
> Stephen Walton <stephen.walton at csun.edu>
> Dept. of Physics & Astronomy, Cal State Northridge
>
> _______________________________________________
> SciPy-user mailing list
> SciPy-user at scipy.net
> http://www.scipy.net/mailman/listinfo/scipy-user
>



From oliphant at ee.byu.edu  Fri May 28 13:49:27 2004
From: oliphant at ee.byu.edu (Travis Oliphant)
Date: Fri, 28 May 2004 11:49:27 -0600
Subject: [SciPy-user] Data convolution.
In-Reply-To: <031E40B9-B062-11D8-B39D-003065F93FF0@cwazy.co.uk>
References: <E237E707-AE37-11D8-9C01-000A95C0BC68@pito.com>
	<40B4E2FD.9040104@ee.byu.edu>	<9E3C17E8-AF62-11D8-BF3D-000A95C0BC68@pito.com>
	<40B521C4.6070306@ee.byu.edu>	<1085694345.6683.12.camel@s123n051.sfo.csun.edu>
	<031E40B9-B062-11D8-B39D-003065F93FF0@cwazy.co.uk>
Message-ID: <40B77BA7.8070901@ee.byu.edu>

Lance Boyle wrote:

> When doing (linear) convolutions (of any dimension) with FFTs, does 
> SciPy take care of aliasing in the "other" domain, usually time domain 
> and spatial domain? This is done by zero-padding and with the 
> so-called overlap-add and overlap-save methods. This is a different 
> issue from edge effects (as I understand the earlier posters) which 
> are Gibbs phenomena.
>

We could add an fft-based convolution pretty easily (and would probably 
be worth it).  But, no boundary effects of using fft's are different 
than Gibbs phenomena.   You can ameliorate the boundary effects using 
various schemes as has been alluded to (which one you choose depends on 
how anxious you are about the boundaries).

-Travis O.