Mailman 3 November 2008 - SciPy-User

IPython TextMate Bundle
by Benjamin J. Racine 13 Nov '08

13 Nov '08

I am sending this forward on behalf of Matt Foster... Be sure to look into pysmell (for completion) as well. >>>>>>>>>>>>>>>>>>>>>> Hi All, A similar mail has already been on the (ipython) users mailing list, so my apologies if you've seen most of this before. I've started working on a TextMate bundle for IPython, based on the info on the Wiki [1], the aim is to produce a BSD licensed bundle which helps to integrate TextMate with IPython. I have set up a project on Github [2] which currently contains: * Some help, which doubles as the README * commands for running the current file / line / section in IPython (via applescript, and Terminal.app) * a basic language grammar for ipythonrc config files. The GitHub page contains the README file which has instructions on how to get GetBundles, which will allow you to install the bundle (but not track changes). Alternatively, if you have git, you can get the bundle using the following commands: cd "~/Library/Application Support/TextMate/Bundles" git clone git://github.com/mattfoster/ipython-tmbundle.git IPython.tmbundle osascript -e 'tell app "TextMate" to reload bundles' GitHub users can fork the project and make their own changes. I'd really love to hear any ideas, suggestions or feature requests people have, and I've been told by Fernando that it's ok to use this list for discussions, provided we prefix mail subjects with [TextMate]. Thanks, Matt [1]: http://ipython.scipy.org/moin/Cookbook/UsingIPythonWithTextMate [2]: http://github.com/mattfoster/ipython-tmbundle/ -- Matt Foster | http://my-mili.eu/matt _______________________________________________ IPython-dev mailing list IPython-dev(a)scipy.org http://lists.ipython.scipy.org/mailman/listinfo/ipython-dev

1 0

suggest to change PREREQUISITES of scipy clearly
by oyster 13 Nov '08

13 Nov '08

for example in scipy version = '0.5.1', the 'PREREQUISITES' section of INSTALL.txt says 'NumPy__ 1.0b1 or newer', but when I install numpy version='1.0.4', and run scipy, I get [code] h:\sap-24\bin\lib\site-packages\scipy\misc\__init__.py:25: DeprecationWarning: ScipyTest is now called NumpyTest; please update your code test = ScipyTest().test RuntimeError: module compiled against version 1000002 of C-API but this version of numpy is 1000009 [/code]

2 1

Recommended SVN version?
by jason-sage＠creativetrax.com 12 Nov '08

12 Nov '08

First of all, thanks for the tremendous work everyone has done! In the Sage project (http://www.sagemath.org), we have a slightly patched version of scipy 0.6 currently. We recently upgraded to numpy 1.2 and would like to match that with an upgrade of scipy. We are using scipy more and more; for example, in our next version out later this week, we switched our floating point and complex matrices to use a numpy/scipy backend for most calculations. Is there a recommended SVN version that we should update to while waiting for 0.7 to be released? We're looking for an SVN revision that is relatively stable. Incidentally, count me into the crowd that would find scipy much more valuable if there were more frequent releases; Sage users in general would be testing the code and giving feedback as well. Also, we noticed the following behavior in our current version of scipy, but only on an OSX 10.5 box. If it's easy, can someone see if the following commands give the spurious result we see below for the inverse matrix with scipy.linalg.inv on an OSX 10.5 box? I'd test it, but the code works on our old scipy on my 32 bit Ubuntu box. sage: import numpy sage: a=numpy.array([[1,2,3],[4,5,6],[7,8,9]],dtype="float64") sage: import scipy sage: import scipy.linalg sage: import numpy.linalg sage: scipy.linalg.det(a) 0.0 sage: scipy.linalg.inv(a) array([[ -4.50359963e+15, 9.00719925e+15, -4.50359963e+15], [ 9.00719925e+15, -1.80143985e+16, 9.00719925e+15], [ -4.50359963e+15, 9.00719925e+15, -4.50359963e+15]]) sage: numpy.linalg.det(a) 6.6613381477509392e-16 sage: numpy.linalg.inv(a) array([[ -4.50359963e+15, 9.00719925e+15, -4.50359963e+15], [ 9.00719925e+15, -1.80143985e+16, 9.00719925e+15], [ -4.50359963e+15, 9.00719925e+15, -4.50359963e+15]]) Thanks, Jason

3 3

scipy on old CPU crashes
by oyster 12 Nov '08

12 Nov '08

hi, all I am using an old AMD Duron CPU with Win2k, which seems does not support SSE/SSE2 When I write [code]] >>> from scipy.integrate import quad >>> quad(lambda e:e,1,5) [/code] or [code] >>> import numpy >>> import scipy >>> import scipy.interpolate >>> x = numpy.arange(10,dtype='float32') * 0.3 >>> y = numpy.cos(x) >>> sp = scipy.interpolate.UnivariateSpline(x,y) [/code] python 2.4/python 2.5 (which are all from www.python.org) crash soon. I searched the internet, and found the reason may be SSE/SSE2 instructions in ATLAS, but I am not sure. I found that there are 3 verisons in numpy-1.2.1-win32-superpack-python2.5.exe(numpy-1.2.1-sse3.exe, numpy-1.2.1-sse2.exe and numpy-1.2.1-nosse.exe) Is there a precompiled scipy that judges nosse/sse/sse2 automatically? or is there a way to change ATLAS only according to my CPU? thanx

2 1

convergence detection in optimize/nonlin.py
by Andrew Hawryluk 12 Nov '08

12 Nov '08

I am experimenting with optimize.broyden3() for solving a multivariable, nonlinear problem. The signature is def broyden3(F, xin, iter=10, alpha=0.4, verbose = False) and it is written to iterate exactly 'iter' times. However, after it converges (to within machine tolerances) it runs into division by zero errors and fails while trying to take the square root of a NaN. I have modified my copy as follows. Original (revision 5067), beginning on line 151: #Gm=Gm+(deltaxm-Gm*deltaFxm)*deltaFxm.T/norm(deltaFxm)**2 updateG(deltaxm-Gmul(deltaFxm),deltaFxm/norm(deltaFxm)**2) Modified version, beginning on line 151 normDelta = norm(deltaFxm) if normDelta == 0.0: break #Gm=Gm+(deltaxm-Gm*deltaFxm)*deltaFxm.T/norm(deltaFxm)**2 updateG(deltaxm-Gmul(deltaFxm),deltaFxm/normDelta**2) All of the routines in optimize/nonlin.py have the same behaviour. Could we add some convergence checking to each of them? Should 'iter' be called 'maxiter' to reflect this? Andrew Hawryluk

2 1

scipy.sparse: read file
by Uwe Schmitt 12 Nov '08

12 Nov '08

Hi, I discovered the save-method which writes a sparse matrix to a file. How can construct a sparse matrix from such a file ? I found no appropriate method. Greetings, Uwe

2 2

Vectorized Approach to Matrix Inversion
by Josh Lawrence 12 Nov '08

12 Nov '08

Hey all, I have a NxNxP array. Let's call it foo. foo[:,:,p] contains a matrix I want to invert (or to solve in the Ax = b fasion). Is there a pure python/scipy way to compute an array bar without loops such that it would be equivalent to the following? import scipy.linalg as la import numpy as np bar = np.zeros((N, N, P)) + 0j for i in range(0,P): bar[:,:,i] = la.inv(foo[:,:,i]) Or in the Ax = b sense (with b an NxP array, and A NxNxP): import scipy.linalg as la import numpy as np x = np.zeros((N, P)) + 0j for i in range(0,P): x[:,i] = la.solve(A[:,:,i], b[:,i]) I realize I could write some fortran code to do this or even use cython, but it would be nice if I could do this without needing to compile some extra code. As a summary, does anyone know how to compute the above (either example, but preferably both) without using loops? Cheers, Josh

1 0

scipy.optimize.leastsq and covariance matrix meaning
by massimo sandal 12 Nov '08

12 Nov '08

Hi, I have a trouble with the covariance matrix in the output of scipy.optimize.leastsq . I am trying to find the estimated sigma of the parameters obtained by the fit. Please bear with me since my statistics knowledge is poor. I understand that the diagonal of the covariance matrix should return me the variance values of each parameter. Problems are: 1) The variance of such parameters look unreasonably large to me, despite the fact I obtain an *excellent* fit over a lot of data points (and values extremly well coherent with expected). 2) The non-diagonal values of the covariance are also unreasonably large, which lets me doubt that picking simply the diagonal values is the correct thing to do. The residuals function is: def residuals(params,y,x,T): ''' Calculates the residuals of the fit ''' lambd, pii=params Kb=(1.38065e-23) therm=Kb*T err = y-( (therm*pii/4) * (((1-(x*lambd))**-2) - 1 + (4*x*lambd)) ) return err For example, a common entity of values is: 4390808.6184609979 3993219683.7749424 and the relative covariance matrix is [[ 1.97019986e+29 -2.67163157e+33] [ -2.67163157e+33 3.78415451e+37]] ...which concerns me. m. -- Massimo Sandal , Ph.D. University of Bologna Department of Biochemistry "G.Moruzzi" snail mail: Via Irnerio 48, 40126 Bologna, Italy email: massimo.sandal(a)unibo.it web: http://www.biocfarm.unibo.it/samori/people/sandal.html tel: +39-051-2094388 fax: +39-051-2094387

3 13

how to stop optimization when using optimize.fmin_l_bfgs_b()
by Wolfgang Meyer 11 Nov '08

11 Nov '08

In the optimization function scipy.optimize.fmin_l_bfgs_b(func, x0, fprime=None, args=(), approx_grad=0, bounds=None, m=10, factr=10000000.0, pgtol=1.0000000000000001e-05, epsilon=1e-08, iprint=-1, maxfun=15000) >>func<< is a function to minimize. Suppose during the optimization process I detect some errors inside >>func<< and want to stop the optimization process, how should I manoeuvre? Maybe return some random values from >>func<<? The same question for >>fprime<<: what should I do to break the optimization process when I realized errors inside fprime? Thanks! -- Wolfgang Meyer

2 1

Plotting vector field + velocity magnitude in background
by wbrevis 11 Nov '08

11 Nov '08

Hello all, I'm trying to plot one of my experimental data using scipy. Until now, all the work I did was using Matlab. For one of my normal data- visualization, I read ASCII or Binary files containing 4 columns: The first contains the x coordinate, the second the y one, and the third and fourth columns the velocity in the x and y directions (u and v), i.e. file= x y u v (ordered in columns). After reading the data in Matlab, I normally do: pcolor(x,y,sqrt(u.^2+v.^2)), in order to visualize in colors the velocity magnitude and then quiver(x,y,u,v) in order to see the associated vectors. I was reading the manual of scipy, including the plotting tools, but I am a bit lost (too much information to start). Can somebody help me with suggestions on how to read data using scipy and the best way to plot (pcolor+quiver)?. What about the function quiver3d of mlab, can be used for 2d representation of a flow field, together with surf (also mlab). Thank you in advance for your help and suggestions W. Brevis

3 2