Mailman 3 June 2007 - SciPy-User

shape problem after flipud
by Dominik Szczerba June 23, 2007

June 23, 2007

Hi, The following trivial codelet does not work as expected: ------------------------------- from scipy import * import copy shape = (256,256) data = zeros(256*256) data.shape = shape print 'old shape', data.shape print data data=flipud(data) data.shape=(256*256,) print 'new shape', data.shape ------------------------------- exiting with an uncomprehensive error: data.shape=(256*256,) AttributeError: incompatible shape for a non-contiguous array If 'flipud' is ommited, it works as expected. I tried via a deepcopy, the problem persists. Why should flipud invalidate 'reshapeability'? What am I doing wrong? Thanks a lot for any hints, - Dominik -- Dominik Szczerba, Ph.D. Computer Vision Lab CH-8092 Zurich http://www.vision.ee.ethz.ch/~domi

7 20

leastsq w/ Jocobian seg fault
by Lin Shao June 23, 2007

June 23, 2007

Hi, I think there's a bug in leastsq when Jacobian is involved -- the full_output option has to be 1 if Dfun is assigned to a user-defined function, otherwise there's a segmentation fault. gdb shows the seg fault happens here: Program received signal SIGSEGV, Segmentation fault. [Switching to Thread -1210287456 (LWP 17643)] 0xb7e2c01a in malloc_usable_size () from /lib/tls/libc.so.6 If I force not to use the thread-local libc, I still got a seg fault: Program received signal SIGSEGV, Segmentation fault. [Switching to Thread 16384 (LWP 17610)] 0xb7e7f1aa in __libc_malloc_pthread_startup () from /lib/libc.so.6 This problem doesn't happen if Jocobian (Dfun) is None. It also doesn't happen when running on Debian x86_64 version. I'm using scipy 0.5.3, Debian i686 2.6.18 kernel. Thanks! --lin

1 0

nonlinear fit with non uniform error?
by massimo sandal June 22, 2007

June 22, 2007

Hi, We have a set of data that we fit to a nonlinear function using scipy.optimize.leastsq that, AFAIK, uses the Levenberg-Marquardt method. Talking with a collegue of another lab, he pointed me that the dataset we fit usually has intrinsically more noise in the first part of the data than the latter. So he fitted by taking into account the non uniform error -that is, instead of using plain chi-square, giving more weight to the distance from points with less intrinsic error. He told me that on Origin there is a function that does it. Is there something similar on scipy? m. -- Massimo Sandal University of Bologna Department of Biochemistry "G.Moruzzi" snail mail: Via Irnerio 48, 40126 Bologna, Italy email: massimo.sandal(a)unibo.it tel: +39-051-2094388 fax: +39-051-2094387

6 18

PDE Solver in SciPy
by Lorenzo Isella June 21, 2007

June 21, 2007

Deal All, I have been using happily for a quite a while the ODE solver in SciPy (I refer to integrate.odeint). I wonder now if there is any PDE solver available for Python (better if it was incorporated into SciPy). After a bit of online search, the best I could find under the SciPy umbrella was fipy: http://www.scipy.org/FiPy?highlight=%28fipy%29 and http://www.ctcms.nist.gov/fipy/index.html Is that all? I am mainly interested in population balance equations for aerosol science applications. Any suggestion here is really welcome. Kind Regards Lorenzo

2 1

Re: [SciPy-user] nonlinear fit with non uniform error?
by Trevis Crane June 21, 2007

June 21, 2007

As an aside, will those of you who are *more* in the know on this topic than the rest of us suggest a good text that has a worthwhile treatment of this subject (as well as other related data analysis/statistical issues)? I'd love to learn more about it, but just jumping on Amazon and picking a book at almost random seems like a good way to waste a lot of money I don't have on books that I don't need, so if you have a favorite reference or text, I'm interested in knowing about it. thanks, trevis -----Original Message----- From: scipy-user-bounces(a)scipy.org [mailto:scipy-user-bounces@scipy.org] On Behalf Of David Huard Sent: Thursday, June 21, 2007 8:09 AM To: SciPy Users List Subject: Re: [SciPy-user] nonlinear fit with non uniform error? Hi, What you have is an heteroscedastic normal distribution (varying variance) describing the residuals. 2007/6/21, Matthieu Brucher <matthieu.brucher(a)gmail.com>: 1)Does this mean that least squares is NOT ok? Yes, LS is _NOT_ OK because it assumes that the distribution (with its parameters) is the same for all errors. I don't remember exactly, but this may be due to ergodicity Well, let's put things in perspective. You can still use ordinary least-squares. Theoretically, this means you're making the assumption that the error mean and variance are fixed and constant. In your case, this is not true and you can consider the LS solution like an approximation. What will happen under this approximation is that large errors on Cy will tend to dominate the residuals, and values in Ay will probably not be fitted optimally. I advise you try it anyway and visually check whether you care about that or not. 2)What does "rescaling" mean in this context? You must change B and C so that : Ay +/- 5 B'y +/- 5 C'y +/- 5 Or maximize the likelihood of a multivariate normal distribution, whose covariance matrix describes your assumption about the heteroscedasticity of the residuals. \Sigma = | \sigma_A^2 0 0 | | 0 \sigma_B^2 0 | | 0 0 \sigma_C^2 | Heteroscedastic likelihood = -n/2 \ln(2\pi) - 1/2 \sum \ln(\sigma_i^2) -1/2 \sum \sigma_i^{-2} (y_{obs} - y_{sim})^2 You might also consider the possibility that your errors are multiplicative rather than additive. In this case, describing the residuals by a lognormal distribution could make more sense. Maximize lognormal likelihood: L=lognormal(y_sim | ln(y_obs), \sigma) Cheers, David Matthieu _______________________________________________ SciPy-user mailing list SciPy-user(a)scipy.org http://projects.scipy.org/mailman/listinfo/scipy-user

3 2

Re: [SciPy-user] Problems building scipy from svn
by Andres Gonzalez-Mancera June 21, 2007

June 21, 2007

I just had the exact same problem trying to install in a new Mac mini Core Duo (32 bits) with numpy 1.0.3. Using numpy from svn fixed the problem! Andres > Message: 1 > Date: Wed, 20 Jun 2007 01:17:09 +0200 > From: "Guillem Borrell Nogueras" <guillemborrell(a)gmail.com> > Subject: Re: [SciPy-user] Problems building scipy from svn > To: "SciPy Users List" <scipy-user(a)scipy.org> > Message-ID: > <835da2a60706191617r417a444dp3a65f5052b395a7a(a)mail.gmail.com> > Content-Type: text/plain; charset="utf-8" > > Thanks! > > It fixes the error in my 64 bit computer but my 32 bit one is still > reluctant to build scipy. I'll do a bit more research. > > On 6/19/07, David Cournapeau <david(a)ar.media.kyoto-u.ac.jp> wrote: > > > > Guillem Borrell Nogueras wrote: > > > Hello > > > > > > I am trying to build scipy from source and I get this error message > > > > > > building extension " scipy.fftpack._fftpack" sources > > > target build/src.linux-x86_64-2.5/_fftpackmodule.c does not exist: > > > Assuming _fftpackmodule.c was generated with "build_src > > > --inplace" command. > > > error: '_fftpackmodule.c' missing > > > > > > > > > This happens for every module, if I disable the fftpack subpackage in > > > Lib/setup.py I get this > > > > > > building extension "scipy.integrate.vode" sources > > > target build/src.linux-x86_64-2.5/vodemodule.c does not exist: > > > Assuming vodemodule.c was generated with "build_src --inplace" > > > command. > > > error: ' vodemodule.c' missing > > > > > > > > > I am able to build every single subpackage. > > > > > > [guillem@sisyphus scipy]$ cd Lib/fftpack/ > > > [guillem@sisyphus fftpack]$ python setup.py build > > > (...) > > > ok > > > > > > > > > I got the same message on two different computers and I can't find any > > > misconfiguration. > > > > > > Any ideas? > > > > > I got the same problem on 64 bits arch with last released numpy. You > > should update to numpy svn, and the problem is solved. > > > > David > > _______________________________________________ > > SciPy-user mailing list > > SciPy-user(a)scipy.org > > http://projects.scipy.org/mailman/listinfo/scipy-user > > -- Andrés González Mancera, Ph.D. Biofluid Mechanics Lab Department of Mechanical Engineering University of Maryland, Baltimore County andres.gonzalez(a)umbc.edu 410-455-3347

1 0

bug in scipy.signal.firwin?
by Reynaldo June 20, 2007

June 20, 2007

Hi In the following line the sum function is the standard python sum function, which doesn't have axis keyword. return h / sum(h,axis=0) I'm using In [56]: scipy.__version__ Out[56]: '0.5.2' In [57]: scipy.__numpy_version__ Out[57]: '1.0.1' on a Gentoo Linux, 2.6.18-gentoo-r4 -- |_|0|_| |_|_|0| |0|0|0|

2 2

windowed filter design
by Reynaldo June 20, 2007

June 20, 2007

Hi fellows, this is my first message to the list. Sorry for my english. I need to make a program that let me implement a windowed filter; with lowpass, bandpass, highpass and stopband configurations. At the university we use matlab and i can implement windowed filter with fir1 allowing me to pass an argument specifying the configuration but since i wanna implement it using python and scipy, how can i make it possible using signal.firwin? Am i missing something? Sincerely, Reynaldo -- |_|0|_| |_|_|0| |0|0|0|

1 0

Debugging memory exhaustion in Python?
by Greg Novak June 20, 2007

June 20, 2007

I've written Python code to calculate a bunch of things for a bunch of simulations. The code goes through about 5GB in 10-100MB chunks. The problem is that Python eventually runs out of memory, consuming (according to top) 3GB. I don't see why it should be doing this--as far as I know I'm not hanging on to any references of anything. I've fooled around with the garbage collector, turning debugging information on and trying to see if it will give me useful info about who or what is still hanging around in memory. I've tried to delete all the user variables but even after this the garbage collector can't free any more memory. What I need is du for python memory, just to get a sense of how/why this is happening. Anyone have suggestions about how to get traction on this? Thanks, Greg

6 9

Problems building scipy from svn
by Guillem Borrell Nogueras June 20, 2007

June 20, 2007

Hello I am trying to build scipy from source and I get this error message building extension "scipy.fftpack._fftpack" sources > target build/src.linux-x86_64-2.5/_fftpackmodule.c does not exist: > Assuming _fftpackmodule.c was generated with "build_src --inplace" > command. > error: '_fftpackmodule.c' missing This happens for every module, if I disable the fftpack subpackage in Lib/setup.py I get this building extension "scipy.integrate.vode" sources > target build/src.linux-x86_64-2.5/vodemodule.c does not exist: > Assuming vodemodule.c was generated with "build_src --inplace" command. > error: 'vodemodule.c' missing I am able to build every single subpackage. [guillem@sisyphus scipy]$ cd Lib/fftpack/ > [guillem@sisyphus fftpack]$ python setup.py build > (...) > ok I got the same message on two different computers and I can't find any misconfiguration. Any ideas? Thanks, guillem

2 2