Mailman 3 April 2002 - NumPy-Discussion

Multiple Linear Regression?
by Jasper Phillips 29 Apr '02

29 Apr '02

I'm helping my wife with programming for her economics thesis, which needs to calculate a "Multiple Linear Regression" on her data. Does anyone know of any (preferably though not necesarrily free) software that can do this? I'm working in Python, but not limited to it as I can relatively freely access other languages. I'm still looking for a library written in Python, but haven't had any luck. My second thought was Matlab, but looking over the Matlab website, I couldn't find anything like this by a name I recognize. It looks like I might be able to construct something out of a combination of Sparse Matrices and Linear Regesstion, or perhaps the stuff for overdetermined Linear Equations? Another option may be LAPACK routines, but I'm not familiar with those. Does anyone here have any experience with this kind of stuff? Is there a better place to ask? I'm about ready to take a shot at writing something myself, but I'd really rather avoid this if it's been done before. -Jasper

4 4

unsigned short support in NumPy
by Sebastian Haase 29 Apr '02

29 Apr '02

Hi all, I'm _very_ new to NumPy. I was interested in using it for our project, where we acquire data from a CCD camera. The Problem: Each pixel in the image is a 16 bit gray value. What I read in the documentation - there is only 8 bit (unsigned integer) support in numpy (or should I say numericarray) Are there plans to add a "unsigned short" (16 bit) support . How much effort would that be. Regards, Sebastian Haase -- _\\|//_ (' O-O ') ------------------------------ooO-(_)-Ooo-------------------------------------- Sebastian Haase University of California, San Francisco (415)502-4316

2 1

testing Numeric.array([0])
by Rick Ree 29 Apr '02

29 Apr '02

Should Numeric.array([0]) test false? This seems counterintuitive, and is not the case for the regular python array module. This recently caused a subtle bug for me when I wanted to find the indices of an array that met a condition. If only the first element met the condition, the result was array([0]) -- a non-empty result that evaluated false. If this is the intended behavior, can someone tell me the reason? thanks, Rick

1 0

ANN: Numarray-0.3.3
by Todd Miller 29 Apr '02

29 Apr '02

Numarray 0.3.3 --------------------------------- Numarray is an array processing package designed to efficiently manipulate large multi-dimensional arrays. Numarray is modelled after Numeric and features c-code generated from python template scripts, the capacity to operate directly on arrays in files, and improved type promotions. Numarray-0.3.3 features improved support for arrays of complex numbers, re-implementing complex types using generated code. In addition to being faster, the new complex ufuncs are better integrated with the numarray type system, so operations between numarrays and complex scalars now work properly. This release also fixes a problem experienced by RedHat Linux users installing numarray from source. WHERE ----------- Numarray-0.3.3 windows executable installers and source code tar ball is here: http://sourceforge.net/project/showfiles.php?group_id=1369 Numarray is hosted by Source Forge in the same project which hosts Numeric: http://sourceforge.net/projects/numpy/ The web page for Numarray information is at: http://stsdas.stsci.edu/numarray/index.html Trackers for Numarray Bugs, Feature Requests, Support, and Patches are at the Source Forge project for NumPy at: http://sourceforge.net/tracker/?group_id=1369 REQUIREMENTS -------------------------- numarray-0.3.3 requires Python 2.0 or greater. AUTHORS, LICENSE ------------------------------ Numarray was written by Perry Greenfield, Rick White, Todd Miller, JC Hsu, Paul Barrett, Phil Hodge at the Space Telescope Science Institute. Thanks go to Jochen Kupper of the University of North Carolina for his work on Numarray and for porting the Numarray manual to TeX format. Numarray is made available under a BSD-style License. See LICENSE.txt in the source distribution for details. -- Todd Miller jmiller(a)stsci.edu

1 0

kdfio, 1.1.1
by cavallo＠kip.uni-heidelberg.de 29 Apr '02

29 Apr '02

hy, here is the url last version of kdfio a khoros/cantata kdf file importer: nothing special, but it seems working now, at least for me;-) You can find it at: http://kdfio.sourceforge.net This is my (very) small contribution to the numerical python: inside i plugged a way to modularize the code (and writing some skeleton semi-automatically) that could speed up a litte bit writing new code. Before to give a full announcement on sourceforge i will wait a little bit, just to see if there are no bugs around. Fell free to use/change/make what you want, thanks to all, antonio cavallo ps. khoros is available at http://www.khoral.com and it is not a free program: there is just a free student version.

1 0

Graphs in numarray?
by Magnus Lie Hetland 25 Apr '02

25 Apr '02

I'm looking at various ways of implementing graphs in Python (beyond simple dict-based stuff -- more performance is needed). kjbuckets looks like a nice alternative, as does the Boost Graph Library (not sure how easy it is to use with Boost.Python) but if numarray is to become a part of the standard library, it could be beneficial to use that... For dense graphs, it makes sense to use an adjacency matrix directly in numarray, I should think. (I haven't implemented many graph algorithms with ufuncs yet, but it seems doable...) For sparse graphs I guess some sort of sparse array implementation would be useful, although the archives indicate that creating such a thing isn't a core part of the numarray project. What do you think -- is it reasonable to use numarray for graph algorithms? Perhaps an additional module with standard graph algorithms would be interesting? (I'm sure I could contribute some if there is any interest...) And -- is there any chance of getting sparse matrices in numarray? -- Magnus Lie Hetland The Anygui Project http://hetland.org http://anygui.org

5 10

Non-numeric arrays?
by Magnus Lie Hetland 25 Apr '02

25 Apr '02

I can't find this in the docs (although I've heard it's mentioned there)... Is support for non-numeric arrays (such as character arrays or object pointer arrays) as in Numeric planned for numarray? (Perhaps even supported? My version might not be themost recent...) And what about subclasses of numeric types? E.g: # numarray >>> class foo(int): pass >>> a = array(map(foo, xrange(10))) [...] TypeError: Expecting a python numeric type, got a foo # Numeric >>> class foo(int): pass >>> a = array(map(foo, xrange(10))) >>> tupe(a[0]) <type 'int'> Neither behaviour seems very helpful -- I guess numarray's is cleaner... (Although in this case I think an object array could have been nice...) -- Magnus Lie Hetland The Anygui Project http://hetland.org http://anygui.org

3 3

qs on NumPy
by Ralf Juengling 20 Apr '02

20 Apr '02

Hi, I did not find a way in Python to check whether a Numeric array instance is a shared array or not. Could you confirm: there is no way. Is there work underway to make Numeric arrays subclassable? Regards, Ralf -- -------------------------------------------------------------------------- Ralf Jüngling Institut für Informatik - Lehrstuhl für Mustererkennung & Bildverarbeitung Georges-Köhler-Allee Gebäude 52 Tel: +49-(0)761-203-8215 79110 Freiburg Fax: +49-(0)761-203-8262 --------------------------------------------------------------------------

1 0

NumPy initiated reference counting
by Ralf Juengling 20 Apr '02

20 Apr '02

I'm currently tinkering with the following problem and what like to hear your suggestions: Within a C module I define a new Python type 'IM' (representing an image). The indexing or slicing facilities of NumPy arrays were tailormade for the manipulation of the internal data of its instances. Thus, I could provide a method 'asarray', which creates a properly typed array object 'a' referring to the data of an IM instance 'im': a = im.asarray() I could use PyArray_FromDimsAndData() to create the array instance. Unfortunately, this wouldn't work, since 'a' would not get notified about the death of 'im'. However, if I could prevent 'im' from being garbage collected before all array instances referring to its data are deleted, it should work. NumPy's array type uses a mechanism to prevent garbage collection of array instances if there are other instances that share data with it. My idea was, to use this mechanism, that is to let the asarray method increment im's reference count and let a->base refer to im. Do you think this is a reliable approach? Thanks, Ralf -- -------------------------------------------------------------------------- Ralf Jüngling Institut für Informatik - Lehrstuhl für Mustererkennung & Bildverarbeitung Georges-Köhler-Allee Gebäude 52 Tel: +49-(0)761-203-8215 79110 Freiburg Fax: +49-(0)761-203-8262 --------------------------------------------------------------------------

1 0

Icc compiled Python
by rob 19 Apr '02

19 Apr '02

There has been some discussion on the FreeBSD Ports list about an Icc compiled Python. Benchmarks much faster than the normal gcc compiled version. I'm wondering if anyone here knows anything about it. The discussion can be accessed via www.geocrawler.org/ FreeBSD/ freebsd-ports. Rob. -- ----------------------------- The Numeric Python EM Project www.pythonemproject.com

1 0