Mailman 3 September 2010 - SciPy-User

[SciPy-user] Using Sphinx with numpy/scipy template
by bfrenay Oct. 1, 2010

Oct. 1, 2010

Hello, I just migrated all my code documentation to the numpy/scipy format. However, I have some problems with sphinx. I have (i) installed sphinx and numpydoc using easyinstall, (ii) inserted the following line in conf.py extensions= ['sphinx.ext.autodoc', 'sphinx.ext.autosummary', 'numpydoc'] and (iii) created a simple, dummy index.rst file reproduced here bellow .. project documentation master file .. automodule:: project.datasets :members: However, I only get errors (reproduced after the message). Does anyone have an idea about what is going on, please ? Thank you, Benoît $ make html sphinx-build -b html -d _build/doctrees . _build/html Running Sphinx v1.0.4 Exception occurred: File "/usr/local/lib/python2.6/dist-packages/numpydoc-0.3.1-py2.6.egg/numpydoc/numpydoc.py", line 136, in get_directive raise RuntimeError("No directive named '%s' found" % name) RuntimeError: No directive named 'function' found The full traceback has been saved in /tmp/sphinx-err-pCQ8Df.log, if you want to report the issue to the developers. Please also report this if it was a user error, so that a better error message can be provided next time. Either send bugs to the mailing list at <http://groups.google.com/group/sphinx-dev/>, or report them in the tracker at <http://bitbucket.org/birkenfeld/sphinx/issues/>. Thanks! make: *** [html] Erreur 1 -- View this message in context: http://old.nabble.com/Using-Sphinx-with-numpy-scipy-template-tp29837454p298… Sent from the Scipy-User mailing list archive at Nabble.com.

2 2

Old meaning of numpy mgrid command
by Dav Clark Oct. 1, 2010

Oct. 1, 2010

Hi all, I have some old code that implemented the equivalent of the make3dgabor matlab command. However, the usage of mgrid seems to have changed. The old way, with some variables subbed in for brevity (hence strange arithmetic with constants): ix, iy, it = np.mgrid[0:1.00001:(1/(10-1.0)), 0:1.00001:(1/(10-1.0)), 0.5] Those first two indices can be converted to the more straightforward complex-number approach, yielding: ix, iy, it = np.mgrid[0:1:10*1j, 0:1:10*1j, 0.5] But the float there gives an error, and it's unclear to me what was intended. Does anyone that knows the old behavior of mgrid know what the 0.5 in the third index position would have done? Thanks! Dav

3 3

Integer dtype comparisons
by Michael Pimmer Sept. 30, 2010

Sept. 30, 2010

Hello, is there a nice way to check whether an object is an Integer? Including all python and numpy Integers? I tried out subclass, but uint is not a subclass of Int: >>> a = numpy.zeros(1,numpy.uint32) >>> aType = type(a[0]) >>> issubclass(aType,types.IntType) False I tried out enumerating, but numpy seems to slightly change datatypes during some calculations, which complicates comparisons: >>> a = numpy.zeros(1,numpy.uint8) >>> aValue = a[0] >>> b = numpy.zeros(1,numpy.int32) >>> bValue = b[0] >>> result = aValue + bValue >>> type(result) <type 'numpy.int32'> >>> numpy.int32 <type 'numpy.int32'> >>> type(result) == numpy.int32 False thanks, Michael

4 3

extracting row from 2D array
by Dan Lussier Sept. 29, 2010

Sept. 29, 2010

I want to extract a row from an 2D array given a criteria that is based on each entry in that row. By way of an example: xyz = np.array([[ 159.278 , 14.7733 , 11.6785 ], [ 158.251 , 1.11666, 4.76546], [ 21.1418 , 84.8988 , 56.5169 ], [ 154.149 , 4.9442 , 131.672 ], [ 31.578 , 11.1078 , 8.47435]]) magnitude_sq = xyz*xyz.sum(axis=1) I want to pull out the rows that satisfy a condition per row like magnitude_sq < 16. I've found and tried np.extract, np.take and np.compress. While np.extract fails to do what I want given that it effectively wants to treat the 2D array was a 1D vector, I moved onto np.compress and can get what I want via: np.compress( magnitude_sq < 16 , xyz, axis=0) What I'm trying to understand is the relationship between these different functions (i.e. which one(s) are more fundamental) as the documentation doesn't really identify how they might relate to each other. I have also posted the question to http://ask.scipy.org/en/topic/50-extracting-entries-from-2d-numpy-array to give that system at try.

2 1

Speeding up odeint for big problems
by Justin Bois Sept. 28, 2010

Sept. 28, 2010

Hello all, I'm using odeint to solve a system of PDEs using the method of lines. This works well for problems in one dimension where there are about ~10^2 coupled ODEs. In 2D, I have ~10^4, and speed is a major issue. Does anyone have any general tips on ways to optimally use odeint to solve large problems using method of lines? Some useful info/questions about my problem: --The Jacobian is dense, even if I use finite differencing for the spatial derivatives. --Using something like PyDSTool might be difficult to implement because I need many of SciPy's handy functions like FFTs, etc. --I'm worried I might be wasting time by passing a large object as an argument to the RHS function. This is out of convenience because it contains useful information about the problem. Does odeint do type conversions on all attributes/methods of the input argument, or only on what is needed? Would paring down the input save time? Any input is greatly appreciated. I'd really like to avoid having to code this up in C. Best, Justin

3 3

port a PDE system from matab to scipy
by Massimo Di Stefano Sept. 28, 2010

Sept. 28, 2010

Hello All, i'm tring to port a matlab code to run under scipy. the code solve a pde system using ode45, the matlab code is like : tmax = 100 ; x_d = 60 ; x_s = 0 ; n = 10 ; C_0 = 10 ; V_0 = 5 ; a = 0.9 ; p = 0.15 ; b = 0.65 ; r = 0.1 ; x=linspace(x_s,x_d,n+2); h=(x_d-x_s)/(n+1); C_x_s = 0 ; C_x_d = 0 ; V_x_s = 0 ; V_x_d = 0 ; CViniz=zeros(2*n,1); CViniz(1)=C_0; CViniz(n+1)=V_0; [t,CV] = ode45('AvvDiffPredRatevar',[0 tmax],CViniz); % where AvvDiffPredRatevar is : function CVprimo = AvvDiffPredRatevar(t,CV) global n h x C_x_s C_x_d V_x_s V_x_d a b p r CVprimo(1:n) = -0.2/h*(CV(1:n)-[C_x_s; CV(1:(n-1))])+0.2/(h^2)*.... ([C_x_s; CV(1:n-1)]-2*CV(1:n)+[CV(2:n); C_x_d])+a*CV(1:n)-b*CV(1:n).*CV((n+1):end); CVprimo(n+1:2*n) = -0.2/h*(CV(n+1:end)-[V_x_s; CV(n+1:end-1)])+0.2/(h^2)*.... ([V_x_s; CV(n+1:end-1)]-2*CV(n+1:end)+[CV(n+2:end); V_x_d])+p*CV(n+1:end)-r*CV(1:n).*CV(n+1:end); CVprimo=CVprimo'; in scipy i start to write : tmax = 100 ; x_d = 60 ; x_s = 0 ; n = 10 ; C_0 = 10 ; V_0 = 5 ; a = 0.9 ; p = 0.15 ; b = 0.65 ; r = 0.1 ; x=linspace(x_s,x_d,n+2); h=(x_d-x_s)/(n+1); C_x_s = 0 ; C_x_d = 0 ; V_x_s = 0 ; V_x_d = 0 ; CViniz=zeros(2*n,1); CViniz(1)=C_0; CViniz(n+1)=V_0; scip now i have to define the "ratefunction" AvvDiffPredRatevar(t,CV) have you any suggestion on how to translate the matlab ratefunciotn code to python ? i suppose i have to write a new python function : def AvvDiffPredRatevar(t,CV) and then i have to pass it to scipy : scipy.integrate.ode(f).set_integrator('dopri5') where "f" is "AvvDiffPredRatevar(t,CV)" make sense for you ? or i'm wrong ? thanks for any hints!!! Best regards, Massimo.

1 0

Re: [SciPy-User] Cannot build SciPy : File contains no section headers.
by Sena Johns Sept. 28, 2010

Sept. 28, 2010

> > Message: 1 > Date: Fri, 24 Sep 2010 10:00:24 +0900 > From: David <david(a)silveregg.co.jp> > Subject: Re: [SciPy-User] Cannot build SciPy : File contains no > section headers. > To: SciPy Users List <scipy-user(a)scipy.org> > Message-ID: <4C9BF828.1070001(a)silveregg.co.jp> > Content-Type: text/plain; charset=ISO-8859-1; format=flowed > > On 09/23/2010 12:43 PM, Sena Johns wrote: > > Hi, > > > > I'm trying to install Numpy and SciPy using this following method > > : > http://www.scipy.org/Installing_SciPy/Linux#head-7ce43956a69ec51c6f2cedd894… > > Most likely, you have an invalid site.cfg file. Could you paste it ? > > cheers, > > David > > Thanks David, I've just checked it and found out I misspelled an entry. Cheers, Sena

1 0

stats.chisquare issues
by Gökhan Sever Sept. 28, 2010

Sept. 28, 2010

Hello, Consider these examples: I[35]: np.histogram(ydata, bins=6) O[35]: (array([4, 1, 3, 0, 0, 1]), array([ 2.8 , 146.33333333, 289.86666667, 433.4 , 576.93333333, 720.46666667, 864. ])) I[36]: np.histogram(ypred, bins=6) O[36]: (array([4, 2, 2, 0, 0, 1]), array([ 22.08895 , 166.34439167, 310.59983333, 454.855275 , 599.11071667, 743.36615833, 887.6216 ])) I[45]: stats.chisquare([4, 1, 3, 0, 0, 1], [4, 2, 2, 0, 0, 1])--------------------------------------------------------------------------- AttributeError Traceback (most recent call last) /home/gsever/Desktop/<ipython console> in <module>() /usr/lib/python2.6/site-packages/scipy/stats/stats.pyc in chisquare(f_obs, f_exp, ddof) 2516 if f_exp is None: 2517 f_exp = array([np.sum(f_obs,axis=0)/float(k)] * len(f_obs),float) -> 2518 f_exp = f_exp.astype(float) 2519 chisq = np.add.reduce((f_obs-f_exp)**2 / f_exp) 2520 return chisq, chisqprob(chisq, k-1-ddof) AttributeError: 'list' object has no attribute 'astype' Here, I expect any scipy function including chisquare should be able to handle lists??? ############################################ This one throws: I[46]: stats.chisquare(np.array([4, 1, 3, 0, 0, 1]), np.array([4, 2, 2, 0, 0, 1])) O[46]: (nan, nan) again I should be aware since the division has 0 in it. after masking: I[47]: a1 = np.ma.masked_equal([4,1,3,0,0,1], 0) I[48]: a2 = np.ma.masked_equal([4,2,2,0,0,1], 0) Further, I[49]: stats.chisquare(a1, a2) O[49]: (1.0, 0.96256577324729631) I[50]: stats.mstats.chisquare(a1, a2) O[50]: (1.0, 0.80125195690120077) p-values differ, is this expected? Thanks. -- Gökhan

3 4

Can't install Scipy on my Mac 10.5.2
by 22Nick22 Yanpei Liu Sept. 27, 2010

Sept. 27, 2010

Hi Guys: During the installation I keep getting this message: "You cannot install scipy 0.8.0 on this volume. scipy requires Python 2.6 from www.python.org to install." This has been confusing me for months since I installed Python 2.7 directly from www.python.org. It should have replaced the Apple's stock Python 2.5... When I type python it returns: Python 2.7 (r27:82508, Jul 3 2010, 20:17:05) [GCC 4.0.1 (Apple Inc. build 5493)] on darwin Any suggestion is appreciated. Thanks Nick

2 1

leastsq and cython
by Till Stensitzki Sept. 27, 2010

Sept. 27, 2010

Hello, i have a problem with scipy.optimize.leastsq and cython. To make my fitting procedure faster and cythonized parts of it and got a factor 2 speedup. But if i try to use the cython version with leastsq i get: AttributeError: 'builtin_function_or_method' object has no attribute 'func_name' any suggestions? I am using scipy 0.8 and cython 0.13 on Windows with Python 2.6.

6 7