Mailman 3 January 2010 - NumPy-Discussion

wired error message in scipy.sparse.eigen function: Segmentation fault
by Jankins Feb. 10, 2010

Feb. 10, 2010

Dear all, I am using scipy '0.8.0.dev6120'. And the scipy.sparse.eigen function always produces error message. _Description:_ linalg.eigen(A, k=6, M=None, sigma=None, which='LM', v0=None, ncv=None, maxiter=None, tol=0, return_eigenvectors=True)_ Error messages:_ When I use this function in the way : linalg.eigen(A, k=2, return_eigenvectors=False) it produces error : *** glibc detected *** python: double free or corruption (!prev) when I use : linalg.eigen(A, k=4, return_eigenvectors=False) or linalg.eigen(A, k=8, return_eigenvectors=False) it produces error : Segmentation fault _My goal:_ "A" is an _unsymmetrical CSR sparse matrix_. What I am trying to do is : 1. find a node 's' to delete. For edge (u, v) in all the out-edges and in-edges of node 's', I set A[u, v] = 0.0. 2. calculate the largest eigenvalue using linalg.eigen(A, return_eigenvectors=False) 3. repeat 1-2 steps many times. I had used eigen_symmetric function to compute symmetric CSR sparse matrix and it works very well. But for the 'eigen' function, it is not working very well. Could you please help me about it? If it does not work, I have to rewrite my code in MATLAB, which is what I am trying to avoid. Thanks so much. Yours sincerely, Jankins

4 9

Compact way of performing array math with specified result type?
by Russell E. Owen Feb. 8, 2010

Feb. 8, 2010

I often find myself doing simple math on sequences of numbers (which might or might not be numpy arrays) where I want the result (and thus the inputs) coerced to a particular data type. I'd like to be able to say: numpy.divide(seq1, seq2, dtype=float) but ufuncs don't allow on to specify a result type. So I do this instead: numpy.array(seq1, dtype=float) / numpy.array(seq2, dtype=float) Is there a more compact solution (without having to create the result array first and supply it as an argument)? -- Russell

4 3

scipy-tickets restarted emailing on jan17 - how about numpy-tickets ?
by Sebastian Haase Feb. 8, 2010

Feb. 8, 2010

Hi, long time ago I had subscript to get both scipy-tickets and numpy-tickets emailed. Now scipy-tickets apparently started emailing again on 17th of Januar. Will numpy-tickets also come back "by itself" - or should I resubscribe? Regards, Sebastian Haase

2 2

which Python for OS X to build installers?
by Ralf Gommers Feb. 1, 2010

Feb. 1, 2010

Hi, With only a few changes (see diff below) to pavement.py I managed to build a dmg installer. For this I used the Python in the bootstrap virtualenv however, instead of the one in /Library/Frameworks/Python.framework/. Does this matter? I don't have a framework build installed, the 4-way universal build did not work for me out of the box a while ago, and installer downloads from python.org are blocked here (thanks Chinese Govt., the Great Firewall is an impressive productivity killer). So I stuck with the default Apple Python till now. For making releases, would I need the framework build? Do I need 32- and 64-bit versions of Python 2.4, 2.5 and 2.6? Cheers, Ralf diff --git a/pavement.py b/pavement.py index f6c1433..bc931ec 100644 --- a/pavement.py +++ b/pavement.py @@ -88,12 +88,13 @@ SUPERPACK_BUILD = 'build-superpack' SUPERPACK_BINDIR = os.path.join(SUPERPACK_BUILD, 'binaries') options(bootstrap=Bunch(bootstrap_dir="bootstrap"), - virtualenv=Bunch(packages_to_install=["sphinx", "numpydoc"], no_site_packages=True), + virtualenv=Bunch(packages_to_install=["sphinx", "numpydoc"], + no_site_packages=False), sphinx=Bunch(builddir="build", sourcedir="source", docroot='doc'), superpack=Bunch(builddir="build-superpack"), installers=Bunch(releasedir="release", installersdir=os.path.join("release", "installers")), - doc=Bunch(doc_root="doc", + doc=Bunch(doc_root="doc", sdir=os.path.join("doc", "source"), bdir=os.path.join("doc", "build"), bdir_latex=os.path.join("doc", "build", "latex"), @@ -106,7 +107,7 @@ options(bootstrap=Bunch(bootstrap_dir="bootstrap"), MPKG_PYTHON = { "2.5": ["/Library/Frameworks/Python.framework/Versions/2.5/bin/python"], - "2.6": ["/Library/Frameworks/Python.framework/Versions/2.6/bin/python"] + "2.6": ["python"] } SSE3_CFG = {'ATLAS': r'C:\local\lib\yop\sse3'} @@ -206,7 +207,7 @@ def bdist_superpack(options): copy_bdist("sse2") bdist_wininst_arch(pyver, 'sse3') copy_bdist("sse3") - + idirs = options.installers.installersdir pyver = options.python_version prepare_nsis_script(pyver, FULLVERSION) @@ -273,8 +274,8 @@ def bootstrap(options): options.virtualenv.script_name = os.path.join(options.bootstrap_dir, bscript) - options.virtualenv.no_site_packages = True - options.bootstrap.no_site_packages = True + options.virtualenv.no_site_packages = False + options.bootstrap.no_site_packages = False call_task('paver.virtual.bootstrap') sh('cd %s; %s %s' % (bdir, sys.executable, bscript))

2 2

Re: [Numpy-discussion] Distutils issue? [migrated]
by Robert Kern Jan. 30, 2010

Jan. 30, 2010

On Fri, Jan 29, 2010 at 18:18, Tom Davis <tom(a)recursivedream.com> wrote: > [This thread has been migrated from distutils-sig until such a time that it > can be determined if this is a distutils or numpy issue] > A quick recap: Basically, trying to > fix http://projects.scipy.org/numpy/ticket/999 along with some duplicates > thereof. Robert made some changes to get us past the initial missing lists > in 8080 and 8081; now things have gotten crazy. > > ``export CFLAGS="-I/usr/include/python2.5"`` fixed the missing header issue. > I've attached the resulting dump with the patched debug output; I wasn't > able to make a lot of sense out of it. At over 2400 lines, it's a bit > lengthy. Don't use CFLAGS. Please show the dump without that flag. Also, please show me the command line(s) that you are using to try to install numpy. -- Robert Kern "I have come to believe that the whole world is an enigma, a harmless enigma that is made terrible by our own mad attempt to interpret it as though it had an underlying truth." -- Umberto Eco

1 0

How to get the shape of an array slice without doing it
by David Huard Jan. 29, 2010

Jan. 29, 2010

Hi, I have a 4D "array" with a given shape, but the array is never actually created since it is large and distributed over multiple binary files. Typical usage would be to take slices across the 4D array. I'd like to know what the shape of the resulting array would be if I took a slice out of it. That is, let's say my 4D array is A, I'd like to know A[ndindex].shape without actually creating A. ndindex should support all numpy constructions (integer, boolean, array, slice, ...). I am guessing something already exists to do this, but I just can't put my finger on it. Thanks. David

3 10

Warning on http://scipy.org/ about binary incompatibility ?
by josef.pktd＠gmail.com Jan. 29, 2010

Jan. 29, 2010

Can we/someone add a warning on the front page http://scipy.org/ (maybe under news for numpy download) about incompatibility of the binaries on sourceforge of scipy <=0.7.1 with numpy 1.4.0 ? It would avoid that users have to find it out for themselves and reduce questions on the mailing list. An thanks a lot to David for hunting this down. Especially for users like me who have to rely on (some) precompiled binary distributions. Josef

5 14

dtype and logical_and
by Keith Goodman Jan. 28, 2010

Jan. 28, 2010

I noticed that & (logical and) does support float dtype: >> b = np.array([True, False]) >> f = np.array([1.0, 2.0]) >> i = np.array([1, 2]) >> b & b array([ True, False], dtype=bool) >> i & i array([1, 2]) >> i & f TypeError: unsupported operand type(s) for &: 'int' and 'float' But this works: >> np.logical_and(b, f) array([ True, False], dtype=bool) and this gives a different result from i & i above: >> np.logical_and(i, i) array([ True, True], dtype=bool) Why are & and np.logical_and different? If I have a class (a labeled array, larry) any suggestions on whether I should use & or np.logical_and on the the underlying arrays for the __and__ method?

2 2

Re: [Numpy-discussion] NumPy-Discussion Digest, Vol 40, Issue 70
by Robert Kiwanuka Jan. 28, 2010

Jan. 28, 2010

[snip] > > Message: 2 > Date: Thu, 28 Jan 2010 12:50:03 +0000 > From: Robert Kiwanuka <robert.kiwanuka(a)gmail.com> > Subject: [Numpy-discussion] is there any alternative to savefig? > To: numpy-discussion(a)scipy.org > Message-ID: > <b1e2a9a21001280450t6e9c9ee1n351ea3dcc86f80ef(a)mail.gmail.com> > Content-Type: text/plain; charset="iso-8859-1" > > Hi all, > > I wonder if anyone knows any alternative function in pylab (or > otherwise) that could be used to save an image. My problem is as > follows: > > --------------- > from pylab import * > ... > > figure(1) > fig1 = gca() > figure(2) > fig2 = gca() > figure(3) > fig3 = gca() > > for i,data_file in enumerate(data_file_list): > time,x, y,x2, y2 = read_csv_file_4(open > (data_file),elements=num_of_ > elements) > fig1.plot(-x,-y,color=colours[i],label=labellist[i]) > fig2.plot(time,-y,color=colours[i],label=labellist[i]) > fig3.plot(time,-x,color=colours[i],label=labellist[i]) > > fig1.legend(loc='best') > fig1.set_title("y1 - x1") > fig1.set_ylabel("y1") > fig1.set_xlabel("x1") > #savefig("y1-x1.png") > > fig2.legend(loc='best') > fig2.set_title("y1 - time") > fig2.set_ylabel("y1") > fig2.set_xlabel("time[s]") > #savefig("y1-time.png") > > fig3.legend(loc='best') > fig3.set_title("x1 - time") > fig3.set_ylabel("x1") > fig3.set_xlabel("time[s]") > #savefig("x1-time.png") > show() > --------------------------- > > In the above code, I read multiple data files and plot three separate > figures. Now I would like to save each of the figures to a file as the > commented savefig satements suggest. The trouble is that if I > uncomment all those savefig statements, I get three saved images all > containing the plot belonging to figure(3), which was the last figure > declared. > > I understand this to be happening because savefig will save the > "current" figure, which in this case happens to be the last one > declared. > > If I could do something like fig1.savefig("y1-x1.png") or savefig("y1- > x1.png").fig1, this would solve the problem but I'm not aware of any > such methods or modules to enable this. This is thus a flaw in the > general design/implementation of the savefig function, but is there an > alternative function to enable me achieve what I need? Is there > perhaps a possible tweak to savefig to make it do the same? > > Thanks in advance, > > Robert >

1 0

[ANN] New open source project for labeled arrays
by Keith Goodman Jan. 28, 2010

Jan. 28, 2010

I recently opened sourced one of my packages. It is a labeled array that I call larry. A two-dimensional larry, for example, contains a 2d NumPy array with labels on each row and column. A larry can have any dimension. Alignment by label is automatic when you add (or subtract, multiply, divide) two larrys. larry has built-in methods such as movingsum, ranking, merge, shuffle, zscore, demean, lag as well as typical NumPy methods like sum, max, std, sign, clip. NaNs are treated as missing data. You can archive larrys in HDF5 format using save and load or using a dictionary-like interface. I'm working towards a 0.1 release. In the meantime, comments, suggestions, critiques are all appreciated. To use larry you need Python and NumPy 1.4 or newer. To save and load larrys in HDF5 format, you need h5py with HDF5 1.8. larry currently contains no extensions, just Python code, so there is nothing to compile. Just save the la package and make sure Python can find it. docs http://larry.sourceforge.net code https://launchpad.net/larry

5 7