Mailman 3 October 2004 - SciPy-User

debian package versions...
by Gerald Richter 22 Oct '04

22 Oct '04

just installed the new debian packages: # dpkg -s python-scipy Package: python-scipy Status: install ok installed Priority: extra Section: python Installed-Size: 44 Maintainer: Jos\uffff\uffff Fonseca <j_r_fonseca(a)yahoo.co.uk> Architecture: all Version: 0.3.2-1 ... # dpkg -s python-scipy-core Package: python-scipy-core Status: install ok installed Priority: extra Section: python Installed-Size: 40 Maintainer: Jos\uffff\uffff Fonseca <j_r_fonseca(a)yahoo.co.uk> Architecture: all Version: 0.3.2-1 ... BUT: $ ipython -p scipy In [1]: scipy.__version__ Out[1]: '0.3.0_266.4239' is this just about forgetting to set the right __version__, or is this a bug in the package composition? -- Gerald Richter <richter(a)hephy.oeaw.ac.at> phone: +43 1 5447328/27 Institute of high energy physics (HEPHY) http://wwwhephy.oeaw.ac.at/ Vienna/Austria

2 2

Sparse Complex matrices and iterative solvers
by Nils Wagner 22 Oct '04

22 Oct '04

Hi all, Iterative solvers cannot be used with sparse complex matrices. Traceback (most recent call last): File "sparse_it.py", line 32, in ? sol0,info0 = linalg.gmres(A,r) File "/usr/lib/python2.3/site-packages/scipy/linalg/iterative.py", line 499, in gmres work[slice2] += sclr1*matvec(work[slice1]) TypeError: return array has incorrect type Nils

1 0

A.todense() for complex sparse matrices
by Nils Wagner 22 Oct '04

22 Oct '04

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 * >>> A = sparse.dok_matrix() >>> A[0,0] = 1j >>> A.todense() Traceback (most recent call last): File "<stdin>", line 1, in ? File "/usr/lib/python2.3/site-packages/scipy/sparse/Sparse.py", line 1257, in todense new[ikey0,ikey1] = self[key] TypeError: can't convert complex to float; use abs(z) >>> Nils

1 0

linalg.norm for sparse matrices
by Nils Wagner 22 Oct '04

22 Oct '04

Hi all, AFAIK, linalg.norm cannot be used for sparse matrices >>> linalg.norm(A) 0.0 >>> linalg.norm(A.todense()) 1.3656824382572492 Nils

1 0

Bugs in linalg.qmr linalg.bicg
by Nils Wagner 22 Oct '04

22 Oct '04

Hi all, I found two bugs in linalg.qmr and linalg.bicg. Traceback (most recent call last): File "sparse_it.py", line 19, in ? sol2,info2 = linalg.qmr(A,r) File "/usr/lib/python2.3/site-packages/scipy/linalg/iterative.py", line 576, in qmr rmatvec = get_rmatvec(A) File "/usr/lib/python2.3/site-packages/scipy/linalg/iterative.py", line 26, in __init__ raise ValueError, "Object must be an array "\ ValueError: Object must be an array or have a callable rmatvec attribute. Traceback (most recent call last): File "sparse_it.py", line 20, in ? sol3,info3 = linalg.bicg(A,r) File "/usr/lib/python2.3/site-packages/scipy/linalg/iterative.py", line 154, in bicg rmatvec = get_rmatvec(A) File "/usr/lib/python2.3/site-packages/scipy/linalg/iterative.py", line 26, in __init__ raise ValueError, "Object must be an array "\ ValueError: Object must be an array or have a callable rmatvec attribute. Nils

1 0

the meaning of c_ and r_
by Joe Harrington 21 Oct '04

21 Oct '04

> And: if I got some functions that might be useful to others, and want to > contribute them, where do I turn to? > I'm working in high-energy physics, and found out, what a beatiful tool for > algorithm prototyping scipy can be... AND I think (and one of my colleagues > knows already ;) it will work even better in data-analysis, together with > the bindings to ROOT (PyROOT) > - http://root.cern.ch > where Python can provide a really reasonable shell and scripting > environment, instead of ROOT's C++ interpreter(!!!), and scipy can do the > necessary data manipulations. If you have or know of software that's not appropriate to add to the core scipy release but that is in coherent packages, please list it on the "Topical Software" wiki on scipy.org. Put your link in the Unindexed section. When there's a critical mass of similar stuff, someone will make a separate page and link it into the index page. If everyone would take a moment to list their favorite packages this way, we'd have a comprehensive listing in no time. --jh--

1 0

New Numeric
by Travis Oliphant 21 Oct '04

21 Oct '04

Heads up to SciPy Users A new Numeric has been released: version 23.6 Hopefully this will fix the problems with different versions of Numeric 23.5 floating around In addition argmin now works for unsigned arrays and another bug that wouldn't have shown up unless you use unsigned integer arrays with elements that are larger than the largest signed integer type was fixed. A windows binary won't be released until later this evening. But, rpm's and tarballs are available now at sourceforge. -Travis O.

1 0

scipy debian packages
by Arnd Baecker 21 Oct '04

21 Oct '04

Hi, I just installed the debs from http://deb-scipy.alioth.debian.org/apt/ (from 18 Oct, giving scipy 3.2). In [1]: import scipy In [2]: scipy.__version__ Out[2]: '0.3.2' and get In [3]: from scipy.xplt import * --------------------------------------------------------------------------- PPImportError Traceback (most recent call last) (full traceback below...) Somehow I thought from the archive that this was fixed, (http://www.scipy.org/mailinglists/mailman?fn=scipy-user/2004-October/003385…) but maybe this is a new/different incarnation? ((I also checked that this also happens with the "normal" python. I am using debian testing.)) Is this also found, e.g., on other systems (windows, fedora, ...)? Arnd Full traceback: In [3]: from scipy.xplt import * --------------------------------------------------------------------------- PPImportError Traceback (most recent call last) /home/abaecker/<console> /usr/lib/python2.3/site-packages/scipy_base/ppimport.py in __getattr__(self, name) 301 module = self.__dict__['_ppimport_module'] 302 except KeyError: --> 303 module = self._ppimport_importer() 304 return getattr(module, name) 305 /usr/lib/python2.3/site-packages/scipy_base/ppimport.py in _ppimport_importer(self) 260 exc_info = self.__dict__.get('_ppimport_exc_info') 261 if exc_info is not None: --> 262 raise PPImportError,\ 263 ''.join(traceback.format_exception(*exc_info)) 264 else: PPImportError: Traceback (most recent call last): File "/usr/lib/python2.3/site-packages/scipy_base/ppimport.py", line 273, in _ppimport_importer module = __import__(name,None,None,['*']) File "/usr/lib/python2.3/site-packages/scipy/xplt/__init__.py", line 12, in ? from Mplot import * File "/usr/lib/python2.3/site-packages/scipy/xplt/Mplot.py", line 1015, in ? import colorbar File "/usr/lib/python2.3/site-packages/scipy/xplt/colorbar.py", line 9, in ? from slice3 import * File "/usr/lib/python2.3/site-packages/scipy/xplt/slice3.py", line 1581, in ? _poly_permutations4 = _construct3 (0) File "/usr/lib/python2.3/site-packages/scipy/xplt/slice3.py", line 1575, in _construct3 mask = find_mask (below, _node_edges3 [itype]) TypeError: Array can not be safely cast to required type

3 7

Event monitor for ode solver?
by Gabriel Gellner 21 Oct '04

21 Oct '04

Is there any mechanism in SciPy for monitoring some relationship to the variables from an call to odeint as they are found? (I'm am thinking of something like the Event mechanism in matlab) I am doing simple bifurcation analysis, harvisting the local max/min values using ODEPACK in fortran 90. But I would like to do some of this type of analysis in SciPy if possible . . . thanks, Gabriel

1 0

vectorize in scipy 0.3.2?
by Francesc Alted 21 Oct '04

21 Oct '04

Hi, I'm having problems with the vectorize function of scipy. Look: $ python Python 2.3.4 (#1, Jul 22 2004, 20:47:54) [GCC 3.3.2 20031022 (Red Hat Linux 3.3.2-1)] on linux2 Type "help", "copyright", "credits" or "license" for more information. >>> from scipy import * >>> def addsubtract(a,b): ... if a > b: ... return a - b ... else: ... return a + b ... >>> vec_addsubstract = vectorize(addsubtract) >>> vec_addsubtract([0,3,6,9],[1,3,5,7]) Traceback (most recent call last): File "<stdin>", line 1, in ? NameError: name 'vec_addsubtract' is not defined and I've tested it on a debian box, with debian packages (http://deb-scipy.alioth.debian.org/apt) with the same result. Some hint? Cheers, -- Francesc Alted

2 2