From rex@apl.washington.edu Fri Dec 3 01:20:26 1999
From: rex@apl.washington.edu (Rex Andrew)
Date: Thu, 2 Dec 1999 17:20:26 -0800 (PST)
Subject: [Matrix-SIG] Q: scientific netcdf crashes python on import?
Message-ID: <Pine.GSO.4.10.9912021709330.25544-100000@kraken.apl.washington.edu>
Hi folks. I've spent about 8 hours on this one, so maybe I could
use some help. I'm trying to install the netCDF module from
ScientificPython on a RH 6.0 linux box. The error is:
Fatal python error: can't initialize module Scientific_netcdf
The crash obviously occurs in NetCDF.py on the line
>>>from Scientific_netcdf import *
OK, not being a guru, I tried (1) loading all the python RPMs and the
new ScientificPython RPM, and/or (2) building both (and including
NumPy source to do so) from source. Several times, in case I wasn't
cleaning up after myself. To no avail.
I tried launching python in gdb, but I'm pretty inept here.
I got python to crash with this error message:
gdb: program received signal SIGABRT Aborted
0x40066111 in __kill() from /lib/libc.so.6
Advice?
-rex "lost in linux land"
-------------------------
Rex K. Andrew randrew@apl.washington.edu
Applied Physics Laboratory 206.543.1250
University of Washington 206.543.6785 (FAX)
Seattle, Washington, USA http://staff.washington.edu/rxandrew
From hinsen@cnrs-orleans.fr Fri Dec 3 15:59:05 1999
From: hinsen@cnrs-orleans.fr (Konrad Hinsen)
Date: Fri, 3 Dec 1999 16:59:05 +0100
Subject: [Matrix-SIG] Q: scientific netcdf crashes python on import?
In-Reply-To: <Pine.GSO.4.10.9912021709330.25544-100000@kraken.apl.washington.edu>
(message from Rex Andrew on Thu, 2 Dec 1999 17:20:26 -0800 (PST))
References: <Pine.GSO.4.10.9912021709330.25544-100000@kraken.apl.washington.edu>
Message-ID: <199912031559.QAA12085@chinon.cnrs-orleans.fr>
> Hi folks. I've spent about 8 hours on this one, so maybe I could
> use some help. I'm trying to install the netCDF module from
> ScientificPython on a RH 6.0 linux box. The error is:
> Fatal python error: can't initialize module Scientific_netcdf
> The crash obviously occurs in NetCDF.py on the line
> >>>from Scientific_netcdf import *
Can you import Numerical Python correctly? Just start Python and type
"import Numeric". If that doesn't work, importing netCDF won't work
either, because it does the equivalent of "import Numeric" internally.
That (i.e. a missing or faulty NumPy installation) is the most
frequent cause for this error message in my experience. Also check if
your netCDF library is working (e.g. by running the ncdump utility).
> OK, not being a guru, I tried (1) loading all the python RPMs and the
> new ScientificPython RPM, and/or (2) building both (and including
> NumPy source to do so) from source. Several times, in case I wasn't
> cleaning up after myself. To no avail.
Which RPMs are you using? My laptop is a RedHat 6.0 system as well,
and it runs everything fine, using the RedHat Python RPMs, the NumPy
RPMs from Oliver Andrich's collection, and the ScientificPython RPMs
from my Web site.
Konrad.
--
-------------------------------------------------------------------------------
Konrad Hinsen | E-Mail: hinsen@cnrs-orleans.fr
Centre de Biophysique Moleculaire (CNRS) | Tel.: +33-2.38.25.55.69
Rue Charles Sadron | Fax: +33-2.38.63.15.17
45071 Orleans Cedex 2 | Deutsch/Esperanto/English/
France | Nederlands/Francais
-------------------------------------------------------------------------------
From Robert.Harrison@pnl.gov Thu Dec 9 22:14:54 1999
From: Robert.Harrison@pnl.gov (Harrison, Robert J)
Date: Thu, 09 Dec 1999 14:14:54 -0800
Subject: [Matrix-SIG] Version 1.0 of Python Interface to Global Arrrays
Message-ID: <D3D015A6FA03D31189D70008C7247D423054FF@pnlmse15.pnl.gov>
The link at the bottom is to an initial Python wrapper around the
Global Arrays library. This is a numerically-oriented, portable,
parallel programming environment including distributed, shared,
multi-dimension arrays with both one-sided and collective
operations, message passing, and interfaces to parallel BLAS and
linear algebra packages. The Numeric extension is used to
handle local data.
For sometime we've been using Python embedded into our
massively parallel computational chemistry code in order to
provide the end-user with the ability to control fairly
high-level capabilities. However, we've become so enamored
with Python that we're considering exposing a lot more
of our C and Fortran APIs. Most of these APIs are specific
to our application (molecular and solid state electronic structure
and molecular dynamics), but the parallel programming tools
may be of wider interest, hence this post.
The GA library (C and Fortran interfaces, and the Python interface)
is freely distributed in the public domain, and, since many projects
rely upon it, it is well supported. It runs on every parallel
computing platform to which we have access including both massively
parallel super-computers and clusters of SMP workstations. GA
inter-operates with any of the MPI, TCGMSG or PVM message passing
libraries (the Python interface has only been tested with MPI and
TCGMSG).
There are many things still to do, including understanding how to
use the multiarray object to provide slicing, but the current
version is sufficiently complete and robust to be useful.
http://www.emsl.pnl.gov:2080/proj/tms/hpcc_actinides/pyGA/
Robert Harrison (robert.harrison@pnl.gov)
From johann@physics.berkeley.edu Mon Dec 13 19:55:54 1999
From: johann@physics.berkeley.edu (Johann Hibschman)
Date: 13 Dec 1999 11:55:54 -0800
Subject: [Matrix-SIG] SWIG with recent NumPy
Message-ID: <mtln6y7gz9.fsf@astron.berkeley.edu>
Hello,
I'm trying to use SWIG to automatically wrap some C routines and
return the results as NumPy arrays (converting from my own pet vector
structure.) I have typemaps that worked with old versions of NumPy,
but now I have had to insert a call to "import_array()" by hand in the
SWIG-generated C module in the module initialization section.
This means (AFAIK) that I have to hand-patch the C every time I run
SWIG. That's bad. Is there any way to get SWIG to insert the needed
"import_array()" call automatically? Also, are there standard NumPy
SWIG typemaps?
Lots of questions, as ususal. ;-)
--Johann
--
Johann Hibschman johann@physics.berkeley.edu
From managan@llnl.gov Mon Dec 13 20:11:31 1999
From: managan@llnl.gov (Rob Managan)
Date: Mon, 13 Dec 1999 12:11:31 -0800
Subject: [Matrix-SIG] SWIG with recent NumPy
In-Reply-To: <mtln6y7gz9.fsf@astron.berkeley.edu>
References: <mtln6y7gz9.fsf@astron.berkeley.edu>
Message-ID: <v04210102b47b02af5882@[128.115.232.99]>
>Hello,
>
>I'm trying to use SWIG to automatically wrap some C routines and
>return the results as NumPy arrays (converting from my own pet vector
>structure.) I have typemaps that worked with old versions of NumPy,
>but now I have had to insert a call to "import_array()" by hand in the
>SWIG-generated C module in the module initialization section.
>
>This means (AFAIK) that I have to hand-patch the C every time I run
>SWIG. That's bad. Is there any way to get SWIG to insert the needed
>"import_array()" call automatically? Also, are there standard NumPy
>SWIG typemaps?
>
>Lots of questions, as ususal. ;-)
>
I was about to start looking into this myself. Could you send me a
copy of the typemaps so I can see if I can use them? I will be
getting to this after the holidays.
I think that what you want for the "import _array()" is the %init %{
... %} option in SWIG. It claims that code you insert in this block
will be added to the initialization routine.
*-*-*-*-*-*-*-*-*-*-**-*-*-*-*-*-*-*-*-*-*-
Rob Managan <mailto://managan@llnl.gov>
LLNL ph: 925-423-0903
P.O. Box 808, L-095 FAX: 925-422-3389
Livermore, CA 94551-0808
From pavlos@gaaros.msrc.sunysb.edu Mon Dec 13 20:38:49 1999
From: pavlos@gaaros.msrc.sunysb.edu (Pavlos Christoforou)
Date: Mon, 13 Dec 1999 15:38:49 -0500 (EST)
Subject: [Matrix-SIG] SWIG with recent NumPy
In-Reply-To: <v04210102b47b02af5882@[128.115.232.99]>
Message-ID: <Pine.LNX.4.04.9912131536020.32332-200000@gaaros.msrc.sunysb.edu>
This message is in MIME format. The first part should be readable text,
while the remaining parts are likely unreadable without MIME-aware tools.
Send mail to mime@docserver.cac.washington.edu for more info.
--830597660-686800328-945117529=:32332
Content-Type: TEXT/PLAIN; charset=US-ASCII
Rob -
I have these, originally from Michael Bell and Michael Cherkassoff with
some modifications to make them work with Numerical recipes. But as Johann
mentioned, you will need to add the import_array() manually.
Pavlos
On Mon, 13 Dec 1999, Rob Managan wrote:
> >Hello,
> >
> >I'm trying to use SWIG to automatically wrap some C routines and
> >return the results as NumPy arrays (converting from my own pet vector
> >structure.) I have typemaps that worked with old versions of NumPy,
> >but now I have had to insert a call to "import_array()" by hand in the
> >SWIG-generated C module in the module initialization section.
> >
> >This means (AFAIK) that I have to hand-patch the C every time I run
> >SWIG. That's bad. Is there any way to get SWIG to insert the needed
> >"import_array()" call automatically? Also, are there standard NumPy
> >SWIG typemaps?
> >
> >Lots of questions, as ususal. ;-)
> >
>
> I was about to start looking into this myself. Could you send me a
> copy of the typemaps so I can see if I can use them? I will be
> getting to this after the holidays.
>
> I think that what you want for the "import _array()" is the %init %{
> ... %} option in SWIG. It claims that code you insert in this block
> will be added to the initialization routine.
>
> *-*-*-*-*-*-*-*-*-*-**-*-*-*-*-*-*-*-*-*-*-
> Rob Managan <mailto://managan@llnl.gov>
> LLNL ph: 925-423-0903
> P.O. Box 808, L-095 FAX: 925-422-3389
> Livermore, CA 94551-0808
>
>
> _______________________________________________
> Matrix-SIG maillist - Matrix-SIG@python.org
> http://www.python.org/mailman/listinfo/matrix-sig
>
--830597660-686800328-945117529=:32332
Content-Type: TEXT/PLAIN; charset=US-ASCII; name="numrec.i"
Content-Transfer-Encoding: BASE64
Content-ID: <Pine.LNX.4.04.9912131538490.32332@gaaros.msrc.sunysb.edu>
Content-Description:
Content-Disposition: attachment; filename="numrec.i"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--830597660-686800328-945117529=:32332--
From ransom@cfa.harvard.edu Mon Dec 13 20:41:36 1999
From: ransom@cfa.harvard.edu (Scott M. Ransom)
Date: Mon, 13 Dec 1999 20:41:36 +0000
Subject: [Matrix-SIG] SWIG with recent NumPy
References: <mtln6y7gz9.fsf@astron.berkeley.edu> <v04210102b47b02af5882@[128.115.232.99]>
Message-ID: <38555A00.18F47189@cfa.harvard.edu>
Rob Managan wrote:
>
I have a working set of typemaps for various array types (more can be
added easily) at teh following location:
ftp://cfa-ftp.harvard.edu/pub/ransom/numpy.i
Good luck,
Scott
--
Scott M. Ransom Address: Harvard-Smithsonian CfA
Phone: (617) 495-4142 60 Garden St. MS 10
email: ransom@cfa.harvard.edu Cambridge, MA 02138
PGP Fingerprint: D2 0E D0 10 CD 95 06 DA EF 78 FE 2B CB 3A D3 53
From johann@physics.berkeley.edu Mon Dec 13 20:46:38 1999
From: johann@physics.berkeley.edu (Johann Hibschman)
Date: 13 Dec 1999 12:46:38 -0800
Subject: [Matrix-SIG] SWIG with recent NumPy
In-Reply-To: Pavlos Christoforou's message of "Mon, 13 Dec 1999 15:38:49 -0500 (EST)"
References: <Pine.LNX.4.04.9912131536020.32332-200000@gaaros.msrc.sunysb.edu>
Message-ID: <mtd7sa7emp.fsf@astron.berkeley.edu>
Pavlos Christoforou writes:
> I have these, originally from Michael Bell and Michael Cherkassoff with
> some modifications to make them work with Numerical recipes. But as Johann
> mentioned, you will need to add the import_array() manually.
I just tried it---Rob's suggestion of the %init directive works.
Somehow I'd missed that in the SWIG docs. Just add the line:
%init %{ import_array(); %}
to the .i file.
In case it's useful, I've placed my raw, ugly typemap code at:
<http://astro.berkeley.edu/~johann/python/typemaps.i>
just in case anyone wants to take a look.
--
Johann Hibschman johann@physics.berkeley.edu
From pavlos@gaaros.msrc.sunysb.edu Mon Dec 13 20:57:26 1999
From: pavlos@gaaros.msrc.sunysb.edu (Pavlos Christoforou)
Date: Mon, 13 Dec 1999 15:57:26 -0500 (EST)
Subject: [Matrix-SIG] SWIG with recent NumPy
In-Reply-To: <mtd7sa7emp.fsf@astron.berkeley.edu>
Message-ID: <Pine.LNX.4.04.9912131556080.32332-100000@gaaros.msrc.sunysb.edu>
On 13 Dec 1999, Johann Hibschman wrote:
> %init %{ import_array(); %}
Thanks a lot Johann.
And my apologies for sending the attachment on the list. It was intented
for Rob ...
Pavlos
From guido@CNRI.Reston.VA.US Tue Dec 14 23:26:34 1999
From: guido@CNRI.Reston.VA.US (Guido van Rossum)
Date: Tue, 14 Dec 1999 18:26:34 -0500
Subject: [Matrix-SIG] Don't forget to register for the Python conference!
Message-ID: <199912142326.SAA00634@eric.cnri.reston.va.us>
We know that the Python conference isn't until the next millennium.
You still have THREE WHOLE WEEKS to register and qualify for the early
bird registration. However, at least one of those weeks you will have
partying and family gatherings on your mind, and when that week's
over, recovery from the partying and gathering will probably take
priority over registering for the conference, and as a result you
might be PAYING FULL PRICE! (The horror!) That is, if your payment
isn't received by January 5, 2000.
So, be smart and register *before* Christmas. That's still more than
ten days -- plenty of time to make travel arrangements, register for
the conference, and present your boss with the bill (in that order).
Our motto, due to Bruce Eckel, is: "Life's better without braces."
Some highlights from the conference program:
- 8 tutorials on topics ranging from JPython to Fnorb;
- a keynote by Open Source evangelist Eric Raymond;
- another by Randy Pausch, father of the Alice Virtual Reality project;
- a separate track for Zope developers and users;
- live demonstrations of important Python applications;
- refereed papers, and short talks on current topics;
- a developers' day where the feature set of Python 2.0 is worked out.
Come and join us at the Key Bridge Marriott in Rosslyn (across the
bridge from Georgetown), January 24-27 in 2000. Make the Python
conference the first conference you attend in the new millennium!
The early bird registration deadline is January 5. More info:
http://www.python.org/workshops/2000-01/
--Guido van Rossum (home page: http://www.python.org/~guido/)
From cgw@fnal.gov Thu Dec 16 19:14:34 1999
From: cgw@fnal.gov (Charles G Waldman)
Date: Thu, 16 Dec 1999 13:14:34 -0600 (CST)
Subject: [Matrix-SIG] bug in array constructor for type 'O'
Message-ID: <14425.14874.269412.634864@buffalo.fnal.gov>
Just noticed this:
>>> a=array(["foo","bar","baz"])
>>> a
array([[f, o, o],
[b, a, r],
[b, a, z]],'c')
This is slightly odd, but I can rationalize it, since each of the
strings "looks like" an array of characters.
But if I try to force the arraytype to "O" (Python Object)
I get:
>>> a=array(["foo","bar","baz"],'O')
>>> a
array([[foo , foo , foo ],
[bar , bar , bar ],
[baz , baz , baz ]],'O')
This just looks like a bug.
From pascucci@cs.utexas.edu Fri Dec 17 23:25:45 1999
From: pascucci@cs.utexas.edu (pascucci)
Date: Fri, 17 Dec 1999 17:25:45 -0600
Subject: [Matrix-SIG] extending an array
References: <Pine.LNX.4.10.9911081105020.10064-100000@us2.mayo.edu>
Message-ID: <385AC679.FF1AE134@cs.utexas.edu>
This is a multi-part message in MIME format.
--------------7F5B61EB2560496D656966A2
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Is it possible to extend an array "in place"?
Reshape does not do that but resize creates always a copy.
Is there something that behaves like a C realloc?
thanks
Valerio
--------------7F5B61EB2560496D656966A2
Content-Type: text/x-vcard; charset=us-ascii;
name="pascucci.vcf"
Content-Transfer-Encoding: 7bit
Content-Description: Card for pascucci
Content-Disposition: attachment;
filename="pascucci.vcf"
begin:vcard
n:Pascucci;Valerio
tel;fax:(419) 735-3586
tel;work:(512) 232-4331
x-mozilla-html:FALSE
adr:;;;;;;
version:2.1
email;internet:pascucci@acm.org
fn:Valerio Pascucci
end:vcard
--------------7F5B61EB2560496D656966A2--
From lee_barford@yahoo.com Sat Dec 18 22:57:46 1999
From: lee_barford@yahoo.com (Lee Barford)
Date: Sat, 18 Dec 1999 14:57:46 -0800 (PST)
Subject: [Matrix-SIG] Additions to RandomArray
Message-ID: <19991218225746.23808.qmail@web3104.mail.yahoo.com>
The RandomArray module is an interface built on top of
the well-known ranlib random number generation
library. However, RandomArray doesn't bring out
interfaces to most of the kinds of random numbers that
ranlib can generate. I needed access to some of them.
So I extended RandomArray to provide all of the random
variables that ranlib generates.
Attached are documentation files (in HTML and raw
ASCII with identical contents) and patches to LLNL
Distribution 11.
I hope that these additions will be included in a
future LLNL Distribution.
--Lee Barford
__________________________________________________
Do You Yahoo!?
Thousands of Stores. Millions of Products. All in one place.
Yahoo! Shopping: http://shopping.yahoo.com
From tschinke@swissonline.ch Sun Dec 19 15:43:35 1999
From: tschinke@swissonline.ch (Vincenzo Tschinke)
Date: Sun, 19 Dec 1999 16:43:35 +0100
Subject: [Matrix-SIG] Problem with arrays of arrays
Message-ID: <l03020900b482a60f0e3d@[195.24.94.138]>
Hi!
I am doing some basic molecular connectivity operation (graphs). Trying to
convert a list of bonds (list of edges) into a connectivity table (for each
vertex [atom], list of connected vertices), by filling an array of arrays, I
got the right or wrong result depending on how I define the initial arrays.
If I define an array of N emtpy arrays as
array_of_empties = N * [[]]
I get the *wrong* result. If I use the more exotic definition
array_of_empties = map ( lambda x : [] , range( N ))
I get the *right* result. Note that after being created, such two arrays
compare as equal.
Please see the listing below.
Is there anything wrong with the first definition, or do we have a
Python bug here?
===========================================================================
Python 1.5.2b1 (#47, Jan 13 1999, 15:14:59) [CW PPC w/GUSI w/MSL]
>>> ee=[[1,2],[1,3],[1,4]] # 3 edges : 3 vert. connected to centr. vertex
>>> ee
[[1, 2], [1, 3], [1, 4]]
>>> aa=4*[[]] # array of four empty arrays
>>> AA=map(lambda x : [], range(4)) # same as above
>>> aa==AA # arrays are equal !!!
1
>>> for p in ee: # build table of connected verteces
... i=p[0]
... j=p[1]
... aa[i-1].append(j)
... aa[j-1].append(i)
... AA[i-1].append(j)
... AA[j-1].append(i)
...
>>> aa # wrong result !!!
[[2, 1, 3, 1, 4, 1], [2, 1, 3, 1, 4, 1], [2, 1, 3, 1, 4, 1], [2, 1, 3, 1, 4, 1]]
>>> AA # correct result
[[2, 3, 4], [1], [1], [1]]
>>> aa==AA
0
>>>
=======================
Vincenzo Tschinke
Kernmattstr. 24
CH-4102 Binningen
Switzerland
+4161 422 1991 (T+F)
tschinke@swissonline.ch
=======================
From mkeller@cybercable.fr Sun Dec 19 22:35:37 1999
From: mkeller@cybercable.fr (Marc Keller)
Date: Sun, 19 Dec 1999 23:35:37 +0100
Subject: [Matrix-SIG] Problem with arrays of arrays
References: <l03020900b482a60f0e3d@[195.24.94.138]>
Message-ID: <385D5DB9.A648E54D@cybercable.fr>
Vincenzo Tschinke wrote:
> Hi!
>
> I am doing some basic molecular connectivity operation (graphs). Trying to
> convert a list of bonds (list of edges) into a connectivity table (for each
> vertex [atom], list of connected vertices), by filling an array of arrays, I
> got the right or wrong result depending on how I define the initial arrays.
>
> If I define an array of N emtpy arrays as
>
> array_of_empties = N * [[]]
>
> I get the *wrong* result. If I use the more exotic definition
>
> array_of_empties = map ( lambda x : [] , range( N ))
>
> I get the *right* result. Note that after being created, such two arrays
> compare as equal.
>
> Please see the listing below.
>
> Is there anything wrong with the first definition, or do we have a
> Python bug here?
>
> ===========================================================================
>
> Python 1.5.2b1 (#47, Jan 13 1999, 15:14:59) [CW PPC w/GUSI w/MSL]
>
> >>> ee=[[1,2],[1,3],[1,4]] # 3 edges : 3 vert. connected to centr. vertex
> >>> ee
> [[1, 2], [1, 3], [1, 4]]
> >>> aa=4*[[]] # array of four empty arrays
> >>> AA=map(lambda x : [], range(4)) # same as above
> >>> aa==AA # arrays are equal !!!
> 1
> >>> for p in ee: # build table of connected verteces
> ... i=p[0]
> ... j=p[1]
> ... aa[i-1].append(j)
> ... aa[j-1].append(i)
> ... AA[i-1].append(j)
> ... AA[j-1].append(i)
> ...
> >>> aa # wrong result !!!
> [[2, 1, 3, 1, 4, 1], [2, 1, 3, 1, 4, 1], [2, 1, 3, 1, 4, 1], [2, 1, 3, 1, 4, 1]]
> >>> AA # correct result
> [[2, 3, 4], [1], [1], [1]]
> >>> aa==AA
> 0
> >>>
>
> =======================
> Vincenzo Tschinke
> Kernmattstr. 24
> CH-4102 Binningen
> Switzerland
> +4161 422 1991 (T+F)
> tschinke@swissonline.ch
> =======================
>
> _______________________________________________
> Matrix-SIG maillist - Matrix-SIG@python.org
> http://www.python.org/mailman/listinfo/matrix-sig
Using N * [[]], you are building a list with 4 references to the same list.
Using map ( lambda x : [] , range( N )), each call to lambda produces a new empty list.
Your exemple in an other way:
>>> aa=[[],[],[],[]]
>>> for p in ee: # build table of connected verteces
... i=p[0]
... j=p[1]
... aa[i-1].append(j)
... aa[j-1].append(i)
...
>>> aa
[[2, 3, 4], [1], [1], [1]]
>>>
From mkeller@cybercable.fr Mon Dec 20 13:28:30 1999
From: mkeller@cybercable.fr (Marc Keller)
Date: Mon, 20 Dec 1999 14:28:30 +0100
Subject: [Matrix-SIG] bug in array constructor for type 'O'
References: <14425.14874.269412.634864@buffalo.fnal.gov>
Message-ID: <385E2EFE.97882F75@cybercable.fr>
Charles G Waldman wrote:
> Just noticed this:
>
> >>> a=array(["foo","bar","baz"])
> >>> a
> array([[f, o, o],
> [b, a, r],
> [b, a, z]],'c')
>
> This is slightly odd, but I can rationalize it, since each of the
> strings "looks like" an array of characters.
>
> But if I try to force the arraytype to "O" (Python Object)
> I get:
>
> >>> a=array(["foo","bar","baz"],'O')
> >>> a
> array([[foo , foo , foo ],
> [bar , bar , bar ],
> [baz , baz , baz ]],'O')
>
> This just looks like a bug.
>
> _______________________________________________
> Matrix-SIG maillist - Matrix-SIG@python.org
> http://www.python.org/mailman/listinfo/matrix-sig
Seems to be a bug...
But there is some logic here: stop the level of structure analysis when
type differ:
For example if
l= >>> l= 'floor', 'bar', 'ba'
>>> b=array(l,'O')
results in
>>> b
array([[floor , floor , floor , floor , floor ],
[bar , bar , bar , bar , bar ],
[ba , ba , ba , ba , ba ]],'O')
with
>>> b=array((None,)+l)
we get
>>> b
array([None , floor , bar , ba ],'O')
>>>
From emmanuel.viennet@lipn.univ-paris13.fr Mon Dec 20 12:49:13 1999
From: emmanuel.viennet@lipn.univ-paris13.fr (Emmanuel Viennet)
Date: Mon, 20 Dec 1999 13:49:13 +0100
Subject: [Matrix-SIG] bug in array constructor for type 'O'
References: <14425.14874.269412.634864@buffalo.fnal.gov> <385E2EFE.97882F75@cybercable.fr>
Message-ID: <385E25C9.981399D@lipn.univ-paris13.fr>
The 'O' typecode is seldom used and obviously needs some
debugging...
While you're at it, the display (repr) code is also wrong:
>>> b
array([None , floor , bar , ba ],'O')
is incorrect: we should get
array([None , 'floor' , 'bar' , 'ba' ],'O')
Emmanuel
--
Emmanuel Viennet: <emmanuel.viennet@lipn.univ-paris13.fr>
http://www-lipn.univ-paris13.fr/~viennet/
Return-Path: <owner-matrix-sig@python.org>
Delivered-To: matrix-sig@dinsdale.python.org
Received: from python.org (parrot.python.org [132.151.1.90])
by dinsdale.python.org (Postfix) with ESMTP id 350081CD16
for <matrix-sig@dinsdale.python.org>; Sat, 18 Dec 1999 18:06:50 -0500 (EST)
Received: from web3105.mail.yahoo.com (web3105.mail.yahoo.com [204.71.202.190])
by python.org (8.9.1a/8.9.1) with SMTP id SAA14480
for <matrix-sig@python.org>; Sat, 18 Dec 1999 18:06:47 -0500 (EST)
Message-ID: <19991218230702.25490.qmail@web3105.mail.yahoo.com>
Received: from [15.255.176.47] by web3105.mail.yahoo.com; Sat, 18 Dec 1999 15:07:02 PST
Date: Sat, 18 Dec 1999 15:07:02 -0800 (PST)
From: Lee Barford <lee_barford@yahoo.com>
To: Python Matrix Sig <matrix-sig@python.org>,
LLNL Distribution Support <support@icf.llnl.gov>
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="0-1804289383-945558422=:25420"
Subject: [Matrix-SIG] Additions to RandomArray
Sender: matrix-sig-admin@python.org
Errors-To: matrix-sig-admin@python.org
X-BeenThere: matrix-sig@python.org
X-Mailman-Version: 1.2 (experimental)
Precedence: bulk
List-Id: SIG for Built-in Matrix Types in Python <matrix-sig.python.org>
--0-1804289383-945558422=:25420
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
The attachments to my previous message didn't make it.
Please forgive this second attempt. If this doesn't
work, I'll look for some other way of submitting them.
--Lee Barford
__________________________________________________
Do You Yahoo!?
Thousands of Stores. Millions of Products. All in one place.
Yahoo! Shopping: http://shopping.yahoo.com
--0-1804289383-945558422=:25420
Content-Type: text/html; name="RandomArray-additions.html"
Content-Description: RandomArray-additions.html
Content-Disposition: inline; filename="RandomArray-additions.html"
<!doctype html public "-//w3c//dtd html 4.0 transitional//en">
<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<meta name="GENERATOR" content="Mozilla/4.6 [en] (X11; I; Linux 2.2.10 i686) [Netscape]">
</head>
<body>
<h1>
Additions to RandomArray module</h1>
<h2>
Lee A. Barford 18 December 1999</h2>
The RandomArray.py module distributed with the LLNL Distribution 11 contains
an interface to portions of the widely used and well tested random number
generation package ranlib. Ranlib is capable of generating floating point
random variables with a wide variety of probability density functions and
some kinds of discrete random variables as well. Random variables with
these nonuniform distributions have uses in diverse areas like Monte Carlo
simulation and the testing of signal processing algorithms. However, RandomArray.py
did not provide interfaces to those ranlib routines. This document describes
functions that have been added to module RandomArray to allow access to
all of the kinds of random variables provided by ranlib.
<h3>
Floating point random arrays</h3>
<i>standard_normal(shape=ReturnFloat)</i>
<br>The <tt>standard_normal()</tt> function returns an array of the specified
shape that contains double precision floating point numbers normally (Gaussian)
distributed with mean zero and variance and standard deviation one. If
no shape is specified, a single number is returned.
<p><i>normal(mean, variance, shape=ReturnFloat)</i>
<br>The <tt>normal()</tt> function returns an array of the specified shape
that contains double precision floating point numbers normally distributed
with the specified mean and variance. If no shape is specified, a single
number is returned.
<p><i>multivariate_normal(mean, covariance) or multivariate_normal(mean,
covariance, leadingAxesShape)</i>
<br>The <tt>multivariate_normal()</tt> function takes a one dimensional
array argument <tt>mean</tt> and a two dimensional array argument <tt>covariance.</tt>
Suppose the shape of <tt>mean</tt> is <tt>(n,). </tt>Then the shape of
<tt>covariance
</tt>must
be <tt>(n,n).</tt> The <tt>multivariate_normal()</tt> function
returns a double precision floating point array. The effect of the <tt>leadingAxesShape</tt>
parameter is:
<ul>
<li>
If no <tt>leadingAxesShape</tt> is specified, then an array with shape
<tt>(n,)</tt> is returned containing a vector of numbers with a multivariate
normal distribution with the specified mean and covariance.</li>
<li>
If <tt>leadingAxesShape</tt> is specified, then an array of such vectors
is returned. The shape of the output is <tt>leadingAxesShape.append((n,)).
</tt>The
leading indices into the output array select a multivariate normal from
the array. The final index selects one number from within the multivariate
normal.</li>
</ul>
In either case, the behavior of <tt>multivariate_normal()</tt> is undefined
if <tt>covariance </tt>is not symmetric and positive definite.
<p><i>exponential(mean, shape=ReturnFloat)</i>
<br>The <tt>exponential() </tt>function returns an array of the specified
shape that contains double precision floating point numbers exponentially
distributed with the specified mean. If no shape is specified, a single
number is returned.
<p><i>beta(a, b, shape=ReturnFloat)</i>
<br>The <tt>beta() </tt>function returns an array of the specified shape
that contains double precision floating point numbers beta distributed
with alpha parameter <tt>a </tt>and beta parameter <tt>b. </tt>If no shape
is specified, a single number is returned.
<p><i>gamma(a, r, shape=ReturnFloat)</i>
<br>The <tt>gamma() </tt>function returns an array of the specified shape
that contains double precision floating point numbers beta distributed
with location parameter <tt>a </tt>and distribution shape parameter <tt>r.
</tt>If no <tt>shape</tt> is specified, a single number is returned.
<p><i>chi_square(df, shape=ReturnFloat)</i>
<br>The <tt>chi_square()</tt> function returns an array of the specified
shape that contains double precision floating point numbers with the chi
square distribution with <tt>df</tt> degrees of freedom. If no shape is
specified, a single number is returned.
<p><i>noncentral_chi_square(df, nonc, shape=ReturnFloat)</i>
<br>The <tt>noncentral_chi_square()</tt> function returns an array of the
specified shape that contains double precision floating point numbers with
the chi square distribution with <tt>df </tt>degrees of freedom and noncentrality
parameter nconc. If no shape is specified, a single number is returned.
<p><i>F(dfn, dfd, shape=ReturnFloat)</i>
<br>The <tt>F()</tt> function returns an array of the specified shape that
contains double precision floating point numbers with the F distribution
with
<tt>dfn</tt> degrees of freedom in the numerator and <tt>dfd</tt>
degrees of freedom in the denominator. If no shape is specified, a single
number is returned.
<p><i>noncentral_F(dfn, dfd, nconc, shape=ReturnFloat)</i>
<br>The <tt>noncentral_F()</tt> function returns an array of the specified
shape that contains double precision floating point numbers with the F
distribution with <tt>dfn</tt> degrees of freedom in the numerator, <tt>dfd</tt>
degrees of freedom in the denominator, and noncentrality parameter <tt>nconc</tt>.
If no shape is specified, a single number is returned.
<h3>
Integer random arrays</h3>
<i>binomial(trials, prob, shape=ReturnInt)</i>
<br>The <tt>binomial()</tt> function returns an array with the specified
shape that contains integer numbers with the binomial distribution with
<tt>trials</tt>
trials and event probability <tt>prob</tt>. In other words, each value
in the returned array is the number of times an event with probability
<tt>prob
</tt>occurred
within <tt>trials </tt>repeated trials. If no shape is specified, a single
number is returned.
<p><i>negative_binomial(trials, prob, shape=ReturnInt)</i>
<br>The <tt>negative_binomial()</tt> function returns an array with the
specified shape that contains integer numbers with the negative binomial
distribution with <tt>trials</tt> trials and event probability <tt>prob</tt>.
If no shape is specified, a single number is returned.
<p><i>poisson(mean, shape=ReturnInt)</i>
<br>The <tt>poisson()</tt> function returns an array with the specified
shape that contains integer numbers with the Poisson distribution with
the specified mean. If no shape is specified, a single number is returned.
<p><i>multinomial(trials, probs) or multinomial(trials, probs, leadingAxesShape)</i>
<br>The <tt>multinomial()</tt> function returns an array with that contains
integer numbers with the multinomial distribution with <tt>trials</tt>
trials and event probabilities given in <tt>probs</tt>. <tt>probs</tt>
must be a one dimensional array. There are <tt>len(probs)+1</tt> events.
<tt>probs[i</tt>] is the probability of the <tt>i</tt>-th event for <tt>0<=i<len(probs)</tt>.
The probability of event <tt>len(probs)</tt> is <tt>1.-Numeric.sum(prob).</tt>
<p>The first form returns an integer array of shape <tt>(len(probs)+1,)</tt>
containing one multinomially distributed vector. The second form returns
an array of shape <tt>(m, n, ..., len(probs)+1)</tt> where <tt>(m, n, ...)</tt>
is <tt>leadingAxesShape</tt>. In this case, each
<tt>output[i,j,...,:]</tt>
is an integer array of shape <tt>(len(prob)+1,)</tt> containing one multinomially
distributed vector..
<h3>
Examples</h3>
Mostof the functions in this package take zero or more distribution specific
parameters plus an optional shape parameter. The shape parameter gives
the shape of the output array:
<blockquote><tt>Python 1.5.1 (#1, Mar 21 1999, 22:49:36) [GCC egcs-2.91.66
19990314/Li on linux-i386</tt>
<br><tt>Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam</tt>
<br><tt>>>> from RandomArray import *</tt>
<br><tt>>>> print standard_normal()</tt>
<br><tt>-0.435568600893</tt>
<br><tt>>>> print standard_normal(5)</tt>
<br><tt>[-1.36134553 0.78617644 -0.45038718 0.18508556 0.05941355]</tt>
<br><tt>>>> print standard_normal((5,2))</tt>
<br><tt>[[ 1.33448863 -0.10125473]</tt>
<br><tt> [ 0.66838062 0.24691346]</tt>
<br><tt> [-0.95092064 0.94168913]</tt>
<br><tt> [-0.23919107 1.89288616]</tt>
<br><tt> [ 0.87651485 0.96400219]]</tt>
<br><tt>>>> print normal(7., 4., (5,2)) % mean=7, std. dev.=4</tt>
<br><tt>[[ 2.66997623 11.65832615]</tt>
<br><tt> [ 6.73916003 6.58162862]</tt>
<br><tt> [ 8.47180378 4.30354905]</tt>
<br><tt> [ 1.35531998 -2.80886841]</tt>
<br><tt> [ 7.07408469 11.39024973]]</tt>
<br><tt>>>> print exponential(10., 5) % mean=10</tt>
<br><tt>[ 18.03347754 7.11702306 9.8587961 32.49231603 28.55408891]</tt>
<br><tt>>>> print beta(3.1, 9.1, 5) % alpha=3.1, beta=9.1</tt>
<br><tt>[ 0.1175056 0.17504358 0.3517828 0.06965593 0.43898219]</tt>
<br><tt>>>> print chi_square(7, 5) % 7 degrees of freedom (dfs)</tt>
<br><tt>[ 11.99046516 3.00741053 4.72235727 6.17056274 8.50756836]</tt>
<br><tt>>>> print noncentral_chi_square(7, 3, 5) % 7 dfs, noncentrality
3</tt>
<br><tt>[ 18.28332138 4.07550335 16.0425396 9.51192093 9.80156231]</tt>
<br><tt>>>> F(5, 7, 5) % 5 and 7 dfs</tt>
<br><tt>array([ 0.24693671, 3.76726145, 0.66883826, 0.59169068, 1.90763224])</tt>
<br><tt>>>> noncentral_F(5, 7, 3., 5) % 5 and 7 dfs, noncentrality 3</tt>
<br><tt>array([ 1.17992553, 0.7500126 , 0.77389943, 9.26798989, 1.35719634])</tt>
<br><tt>>>> binomial(32, .5, 5) % 32 trials, prob of an event = .5</tt>
<br><tt>array([12, 20, 21, 19, 17])</tt>
<br><tt>>>> negative_binomial(32, .5, 5) % 32 trials: prob of an event
= .5</tt>
<br><tt>array([21, 38, 29, 32, 36])</tt></blockquote>
Two functions that return generate multivariate random numbers (that is,
random vectors with some known relationship between the elements of each
vector, defined by the distribution). They are <tt>multivariate_normal()</tt>
and <tt>multinomial().</tt> For these two functions, the lengths of the
leading axes of the output may be specified. The length of the last axis
is determined by the length of some other parameter.
<blockquote><tt>>>> from Numeric import *</tt>
<br><tt>>>> multivariate_normal([1,2], [[1,2],[2,1]], [2,3])</tt>
<br><tt>array([[[ 0.14157988, 1.46232224],</tt>
<br><tt> [-1.11820295, -0.82796288],</tt>
<br><tt> [ 1.35251635, -0.2575901
]],</tt>
<br><tt> [[-0.61142141, 1.0230465 ],</tt>
<br><tt> [-1.08280948, -0.55567217],</tt>
<br><tt> [ 2.49873002, 3.28136372]]])</tt>
<br><tt>>>> x = multivariate_normal([10,100], [[1,2],[2,1]], 10000)</tt>
<br><tt>>>> x_mean = sum(x)/10000</tt>
<br><tt>>>> print x_mean</tt>
<br><tt>[ 9.98599893 100.00032416]</tt>
<br><tt>>>> x_minus_mean = x - x_mean</tt>
<br><tt>>>> cov = matrixmultiply(transpose(x_minus_mean), x_minus_mean)
/ 9999.</tt>
<br><tt>>>> cov</tt>
<br><tt>array([[ 2.01737122, 1.00474408],</tt>
<br><tt>[ 1.00474408, 2.0009806 ]])</tt></blockquote>
The a priori probabilities for a multinomial distribution must sum to one.
The prior probability argument to <tt>multinomial()</tt> doesn't give the
prior probability of the last event: it is computed to be one minus the
sum of the others.
<blockquote><tt>>>> multinomial(16, [.1, .4, .2]) % prior probabilities
[.1, .4, .2, .3]</tt>
<br><tt>array([2, 7, 1, 6])</tt>
<br><tt>>>> multinomial(16, [.1, .4, .2], [2,3]) % output shape [2,3,4]</tt>
<br><tt>array([[[ 1, 9, 1, 5],</tt>
<br><tt> [ 0, 10, 3, 3],</tt>
<br><tt> [ 4, 9, 3, 0]],</tt>
<br><tt> [[ 1, 6, 1, 8],</tt>
<br><tt> [ 3, 4, 5, 4],</tt>
<br><tt> [ 1, 5, 2, 8]]])</tt></blockquote>
</body>
</html>
--0-1804289383-945558422=:25420
Content-Type: text/plain; name="RandomArray-additions.txt"
Content-Description: RandomArray-additions.txt
Content-Disposition: inline; filename="RandomArray-additions.txt"
Additions to RandomArray module
Lee A. Barford 18 December 1999
The RandomArray.py module distributed with the LLNL Distribution 11
contains an interface to portions of the widely used and well tested
random number generation package ranlib. Ranlib is capable of
generating floating point random variables with a wide variety of
probability density functions and some kinds of discrete random
variables as well. Random variables with these nonuniform
distributions have uses in diverse areas like Monte Carlo simulation
and the testing of signal processing algorithms. However,
RandomArray.py did not provide interfaces to those ranlib
routines. This document describes functions that have been added to
module RandomArray to allow access to all of the kinds of random
variables provided by ranlib.
Floating point random arrays
standard_normal(shape=ReturnFloat)
The standard_normal() function returns an array of the specified shape
that contains double precision floating point numbers normally
(Gaussian) distributed with mean zero and variance and standard
deviation one. If no shape is specified, a single number is returned.
normal(mean, variance, shape=ReturnFloat)
The normal() function returns an array of the specified shape that
contains double precision floating point numbers normally distributed
with the specified mean and variance. If no shape is specified, a
single number is returned.
multivariate_normal(mean, covariance) or multivariate_normal(mean, covariance, leadingAxesShape)
The multivariate_normal() function takes a one dimensional array
argument mean and a two dimensional array argument covariance. Suppose
the shape of mean is (n,). Then the shape of covariance must be
(n,n). The multivariate_normal() function returns a double precision
floating point array. The effect of the leadingAxesShape parameter is:
If no leadingAxesShape is specified, then an array with shape (n,) is
returned containing a vector of numbers with a multivariate normal
distribution with the specified mean and covariance.
If leadingAxesShape is specified, then an array of such vectors is
returned. The shape of the output is
leadingAxesShape.append((n,)). The leading indices into the output
array select a multivariate normal from the array. The final index
selects one number from within the multivariate normal. In either
case, the behavior of multivariate_normal() is undefined if covariance
is not symmetric and positive definite.
exponential(mean, shape=ReturnFloat)
The exponential() function returns an array of the specified shape
that contains double precision floating point numbers exponentially
distributed with the specified mean. If no shape is specified, a
single number is returned.
beta(a, b, shape=ReturnFloat)
The beta() function returns an array of the specified shape that
contains double precision floating point numbers beta distributed with
alpha parameter a and beta parameter b. If no shape is specified, a
single number is returned.
gamma(a, r, shape=ReturnFloat)
The gamma() function returns an array of the specified shape that
contains double precision floating point numbers beta distributed with
location parameter a and distribution shape parameter r. If no shape
is specified, a single number is returned.
chi_square(df, shape=ReturnFloat)
The chi_square() function returns an array of the specified shape that
contains double precision floating point numbers with the chi square
distribution with df degrees of freedom. If no shape is specified, a
single number is returned.
noncentral_chi_square(df, nonc, shape=ReturnFloat)
The noncentral_chi_square() function returns an array of the specified
shape that contains double precision floating point numbers with the
chi square distribution with df degrees of freedom and noncentrality
parameter nconc. If no shape is specified, a single number is
returned.
F(dfn, dfd, shape=ReturnFloat)
The F() function returns an array of the specified shape that contains
double precision floating point numbers with the F distribution with
dfn degrees of freedom in the numerator and dfd degrees of freedom in
the denominator. If no shape is specified, a single number is
returned.
noncentral_F(dfn, dfd, nconc, shape=ReturnFloat)
The noncentral_F() function returns an array of the specified shape
that contains double precision floating point numbers with the F
distribution with dfn degrees of freedom in the numerator, dfd degrees
of freedom in the denominator, and noncentrality parameter nconc. If
no shape is specified, a single number is returned.
Integer random arrays
binomial(trials, prob, shape=ReturnInt)
The binomial() function returns an array with the specified shape that
contains integer numbers with the binomial distribution with trials
trials and event probability prob. In other words, each value in the
returned array is the number of times an event with probability prob
occurred within trials repeated trials. If no shape is specified, a
single number is returned.
negative_binomial(trials, prob, shape=ReturnInt)
The negative_binomial() function returns an array with the specified
shape that contains integer numbers with the negative binomial
distribution with trials trials and event probability prob. If no
shape is specified, a single number is returned.
poisson(mean, shape=ReturnInt)
The poisson() function returns an array with the specified shape that
contains integer numbers with the Poisson distribution with the
specified mean. If no shape is specified, a single number is returned.
multinomial(trials, probs) or multinomial(trials, probs, leadingAxesShape)
The multinomial() function returns an array with that contains integer
numbers with the multinomial distribution with trials trials and event
probabilities given in probs. probs must be a one dimensional
array. There are len(probs)+1 events. probs[i] is the probability of
the i-th event for 0<=i<len(probs). The probability of event
len(probs) is 1.-Numeric.sum(prob). The first form returns an integer
array of shape (len(probs)+1,) containing one multinomially
distributed vector. The second form returns an array of shape (m, n,
..., len(probs)+1) where (m, n, ...) is leadingAxesShape. In this
case, each output[i,j,...,:] is an integer array of shape
(len(prob)+1,) containing one multinomially distributed vector..
Examples
Most of the functions in this package take zero or more distribution
specific parameters plus an optional shape parameter. The shape
parameter gives the shape of the output array:
Python 1.5.1 (#1, Mar 21 1999, 22:49:36) [GCC egcs-2.91.66 19990314/Li on linux-i386
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> from RandomArray import *
>>> print standard_normal()
-0.435568600893
>>> print standard_normal(5)
[-1.36134553 0.78617644 -0.45038718 0.18508556 0.05941355]
>>> print standard_normal((5,2))
[[ 1.33448863 -0.10125473]
[ 0.66838062 0.24691346]
[-0.95092064 0.94168913]
[-0.23919107 1.89288616]
[ 0.87651485 0.96400219]]
>>> print normal(7., 4., (5,2)) % mean=7, std. dev.=4
[[ 2.66997623 11.65832615]
[ 6.73916003 6.58162862]
[ 8.47180378 4.30354905]
[ 1.35531998 -2.80886841]
[ 7.07408469 11.39024973]]
>>> print exponential(10., 5) % mean=10
[ 18.03347754 7.11702306 9.8587961 32.49231603 28.55408891]
>>> print beta(3.1, 9.1, 5) % alpha=3.1, beta=9.1
[ 0.1175056 0.17504358 0.3517828 0.06965593 0.43898219]
>>> print chi_square(7, 5) % 7 degrees of freedom (dfs)
[ 11.99046516 3.00741053 4.72235727 6.17056274 8.50756836]
>>> print noncentral_chi_square(7, 3, 5) % 7 dfs, noncentrality 3
[ 18.28332138 4.07550335 16.0425396 9.51192093 9.80156231]
>>> F(5, 7, 5) % 5 and 7 dfs
array([ 0.24693671, 3.76726145, 0.66883826, 0.59169068, 1.90763224])
>>> noncentral_F(5, 7, 3., 5) % 5 and 7 dfs, noncentrality 3
array([ 1.17992553, 0.7500126 , 0.77389943, 9.26798989, 1.35719634])
>>> binomial(32, .5, 5) % 32 trials, prob of an event = .5
array([12, 20, 21, 19, 17])
>>> negative_binomial(32, .5, 5) % 32 trials: prob of an event = .5
array([21, 38, 29, 32, 36])
Two functions that return generate multivariate random numbers (that is, random vectors with some known relationship between the elements of each vector, defined by the distribution). They are multivariate_normal() and multinomial(). For these two functions, the lengths of the leading axes of the output may be specified. The length of the last axis is determined by the length of some other parameter.
>>> from Numeric import *
>>> multivariate_normal([1,2], [[1,2],[2,1]], [2,3])
array([[[ 0.14157988, 1.46232224],
[-1.11820295, -0.82796288],
[ 1.35251635, -0.2575901 ]],
[[-0.61142141, 1.0230465 ],
[-1.08280948, -0.55567217],
[ 2.49873002, 3.28136372]]])
>>> x = multivariate_normal([10,100], [[1,2],[2,1]], 10000)
>>> x_mean = sum(x)/10000
>>> print x_mean
[ 9.98599893 100.00032416]
>>> x_minus_mean = x - x_mean
>>> cov = matrixmultiply(transpose(x_minus_mean), x_minus_mean) / 9999.
>>> cov
array([[ 2.01737122, 1.00474408],
[ 1.00474408, 2.0009806 ]])
The a priori probabilities for a multinomial distribution must sum to
one. The prior probability argument to multinomial() doesn't give the
prior probability of the last event: it is computed to be one minus
the sum of the others.
>>> multinomial(16, [.1, .4, .2]) % prior probabilities [.1, .4, .2, .3]
array([2, 7, 1, 6])
>>> multinomial(16, [.1, .4, .2], [2,3]) % output shape [2,3,4]
array([[[ 1, 9, 1, 5],
[ 0, 10, 3, 3],
[ 4, 9, 3, 0]],
[[ 1, 6, 1, 8],
[ 3, 4, 5, 4],
[ 1, 5, 2, 8]]])
--0-1804289383-945558422=:25420
Content-Type: application/x-unknown; name="RandomArray.patch"
Content-Transfer-Encoding: base64
Content-Description: RandomArray.patch
Content-Disposition: attachment; filename="RandomArray.patch"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--0-1804289383-945558422=:25420--
From heather@v1.wustl.edu Tue Dec 21 20:15:53 1999
From: heather@v1.wustl.edu (Heather Drury)
Date: Tue, 21 Dec 1999 14:15:53 -0600 (CST)
Subject: [Matrix-SIG] ValueError: object too deep for desired array (??)
Message-ID: <199912212015.OAA01848@v1.wustl.edu>
Hi,
I'm getting the following error on trying to do an array copy:
File "volume.py", line 50, in __init__
self.VoxData3D [:, :, :] = data.astype ('b')
ValueError: Object too deep for desired array
The relevant code is shown below ("data" is the input array",
which in this case is 256x256x256):
self.xdim = data.shape [2]
self.ydim = data.shape [1]
self.zdim = data.shape [0]
self.VoxData3D = zeros ((self.zdim, self.ydim, self.xdim), 'b')
self.VoxData3D [:, :, :] = data.astype ('b')
self.VoxData = ravel (self.VoxData3D)
I can't seem to find anything in the doc about this...and since I'm
allocating the space with the "zeros" command, I don't understand
why the object is "too deep".
Any ideas?
Heather
--
Heather Drury heather@v1.wustl.edu
Washington University School of Medicine http://v1.wustl.edu
Department of Anatomy & Neurobiology Phone: 314-362-4325
660 S. Euclid, MS 8108 FAX: 314-747-4370
St. Louis, MO 63110-1093
From guido@CNRI.Reston.VA.US Wed Dec 22 20:00:19 1999
From: guido@CNRI.Reston.VA.US (Guido van Rossum)
Date: Wed, 22 Dec 1999 15:00:19 -0500
Subject: [Matrix-SIG] Two weeks Till Python Conference Early Bird Registration Deadline!
Message-ID: <199912222000.PAA17264@eric.cnri.reston.va.us>
We know that the Python conference isn't until the next millennium.
You have exactly two weeks left to register and qualify for the early
bird registration. Since most of that time most people are taking off
for the holidays, it's really NOW OR NEVER! If you haven't registered
and paid by January 5, you will paying full price... So, be smart and
register NOW. Also don't forget to book your hotel room by January 3.
Some highlights from the conference program:
- 8 tutorials on topics ranging from JPython to Fnorb;
- a keynote by Open Source evangelist Eric Raymond;
- another by Randy Pausch, father of the Alice Virtual Reality project;
- a separate track for Zope developers and users;
- live demonstrations of important Python applications;
- refereed papers, and short talks on current topics;
- a developers' day where the feature set of Python 2.0 is worked out.
Our motto, due to Bruce Eckel, is: "Life's better without braces."
Come and join us at the Key Bridge Marriott in Rosslyn (across the
bridge from Georgetown), January 24-27 in 2000. Make the Python
conference the first conference you attend in the new millennium!
The early bird registration deadline is January 5. More info:
http://www.python.org/workshops/2000-01/
The program is now complete with the titles of all presentations.
There is still space in the demo session and in the short talks
session.
--Guido van Rossum (home page: http://www.python.org/~guido/)
From crabb2@llnl.gov Thu Dec 23 00:28:54 1999
From: crabb2@llnl.gov (Charlie Crabb)
Date: Wed, 22 Dec 1999 16:28:54 -0800
Subject: [Matrix-SIG] Turning off the traceback on a raise
Message-ID: <38616CC6.2581E05C@glacier.llnl.gov>
Is it possible to turn off the stack trace in python
after raising an exception?
e.g disable output from the first two lines:
Traceback (innermost last):
File "<stdin>", line 4, in ?
KeyError: nada
--
Charlie Crabb
Computer Scientist, LCPD/A/X Divisions
Lawrence Livermore National Laboratory, L-38
crabb2@llnl.gov 925-424-3265
From Oliphant.Travis@mayo.edu Tue Dec 28 22:49:25 1999
From: Oliphant.Travis@mayo.edu (Travis Oliphant)
Date: Tue, 28 Dec 1999 16:49:25 -0600 (CST)
Subject: [Matrix-SIG] New Cephes module with arraymap method
Message-ID: <Pine.LNX.4.10.9912281627520.6360-100000@us2.mayo.edu>
I'm announcing the availability of Version 1.2 of the Cephes module at
http://oliphant.netpedia.net
Cephes module provides many special functions of physics to the Numerical
Python user as ufuncs which are generalized functions that automatically
map over array arguments and follow useful broadcasting rules.
I would like to see more special functions added to the library even if
they are written in Python.
Towards that end, version 1.2 contains a new python wrapper module called
SpecialFuncs.py that can include Python-only special functions.
In addition, I have added a method to cephesmodule called arraymap that
acts like the map function but performs the broadcasting rules of ufuncs
and returns Numerical Python arrays as output.
I have also included a class that allows wrapping a Python function that
takes scalar inputs and returns scalar outputs and returns a ufunc-like
function which automatically maps over array input arguments following the
ufunc broadcasting rules to return appropriately-sized outputs.
So, if you can't figure out how to "vectorize" your function. You can
wrap it in this class and get mapping behavior automatically. The
looping is all written in C and so is reasonably fast.
Comments and contributions are welcome.
Travis Oliphant
Oliphant.Travis@altavista.net
<P><A HREF="http://oliphant.netpedia.net/">SpecialFuncs 1.2</A> - an
extension module also known as cephes which provides many special
functions of physics to the Numerical Python user. (29-Dec-99)
From avoinov@exelixis.com Wed Dec 29 22:16:36 1999
From: avoinov@exelixis.com (Alexander V. Voinov)
Date: Wed, 29 Dec 1999 14:16:36 -0800
Subject: [Matrix-SIG] import Numeric: add.reduce not found
Message-ID: <386A8844.4F3D46BA@exelixis.com>
Hi All
I tried to use Python-1.5.2 with LLNLNumeric11 on Solaris .2.5.1, and it
says:
Python 1.5.2 (#7, Dec 29 1999, 13:03:01) [GCC 2.95 19990728 (release)]
on sunos5
Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
>>> import Numeric
Traceback (innermost last):
File "<stdin>", line 1, in ?
File
"/tools/Python-1.5.2/lib/python1.5/site-packages/numeric/Numeric.py",
line 319, in ?
sum = add.reduce
NameError: add
Thank you in advance
Alexander
From avoinov@exelixis.com Wed Dec 29 23:39:31 1999
From: avoinov@exelixis.com (Alexander V. Voinov)
Date: Wed, 29 Dec 1999 15:39:31 -0800
Subject: [Matrix-SIG] Re: [Fwd: import Numeric: add.reduce not found]
References: <386A9008.8D070112@exelixis.com>
Message-ID: <386A9BB3.B78E56ED@exelixis.com>
"Alexander V. Voinov" wrote:
> I tried to use Python-1.5.2 with LLNLNumeric11 on Solaris .2.5.1, and it
> says:
>
> Python 1.5.2 (#7, Dec 29 1999, 13:03:01) [GCC 2.95 19990728 (release)]
> on sunos5
> Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
> >>> import Numeric
> Traceback (innermost last):
> File "<stdin>", line 1, in ?
> File
> "/tools/Python-1.5.2/lib/python1.5/site-packages/numeric/Numeric.py",
> line 319, in ?
> sum = add.reduce
> NameError: add
The same with LL*10.
Alexander
From avoinov@exelixis.com Thu Dec 30 00:08:14 1999
From: avoinov@exelixis.com (Alexander V. Voinov)
Date: Wed, 29 Dec 1999 16:08:14 -0800
Subject: [Matrix-SIG] Re: [Fwd: import Numeric: add.reduce not found]
References: <386A9008.8D070112@exelixis.com> <386A9BB3.B78E56ED@exelixis.com>
Message-ID: <386AA26E.F4ED73D2@exelixis.com>
"Alexander V. Voinov" wrote:
> "Alexander V. Voinov" wrote:
>
> > I tried to use Python-1.5.2 with LLNLNumeric11 on Solaris .2.5.1, and it
> > says:
> >
> > Python 1.5.2 (#7, Dec 29 1999, 13:03:01) [GCC 2.95 19990728 (release)]
> > on sunos5
> > Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
> > >>> import Numeric
> > Traceback (innermost last):
> > File "<stdin>", line 1, in ?
> > File
> > "/tools/Python-1.5.2/lib/python1.5/site-packages/numeric/Numeric.py",
> > line 319, in ?
> > sum = add.reduce
> > NameError: add
I had to compile everything form Numeric/Setup statically into the 'python'
binary to make it work. It's ok for a short while, but I'd be very grateful
if somebody pointed me to the source of the problem.
Alexander
From jcollins@shell.pacificnet.net Thu Dec 30 17:47:51 1999
From: jcollins@shell.pacificnet.net (Jeff Collins)
Date: Thu, 30 Dec 1999 09:47:51 -0800 (PST)
Subject: [Matrix-SIG] Re: [Fwd: import Numeric: add.reduce not found]
In-Reply-To: <386A9BB3.B78E56ED@exelixis.com>
Message-ID: <Pine.GSO.4.10.9912300944470.27532-100000@shell.pacificnet.net>
Just out of curiosity - do you have an old umathmodule.so
(or old version of NumPy) lying around in your path? If so,
removing it could solve your problem.
Jeff
On Wed, 29 Dec 1999, Alexander V. Voinov wrote:
> "Alexander V. Voinov" wrote:
>
> > I tried to use Python-1.5.2 with LLNLNumeric11 on Solaris .2.5.1, and it
> > says:
> >
> > Python 1.5.2 (#7, Dec 29 1999, 13:03:01) [GCC 2.95 19990728 (release)]
> > on sunos5
> > Copyright 1991-1995 Stichting Mathematisch Centrum, Amsterdam
> > >>> import Numeric
> > Traceback (innermost last):
> > File "<stdin>", line 1, in ?
> > File
> > "/tools/Python-1.5.2/lib/python1.5/site-packages/numeric/Numeric.py",
> > line 319, in ?
> > sum = add.reduce
> > NameError: add
>
> The same with LL*10.
>
> Alexander
>
>
>
> _______________________________________________
> Matrix-SIG maillist - Matrix-SIG@python.org
> http://www.python.org/mailman/listinfo/matrix-sig
>
From Oliphant.Travis@mayo.edu Thu Dec 30 18:32:53 1999
From: Oliphant.Travis@mayo.edu (Travis Oliphant)
Date: Thu, 30 Dec 1999 12:32:53 -0600 (CST)
Subject: [Matrix-SIG] Problems with Numeric
In-Reply-To: <19991230170019.3ABAE1CD7A@dinsdale.python.org>
Message-ID: <Pine.LNX.4.10.9912301228400.10812-100000@us2.mayo.edu>
> I had to compile everything form Numeric/Setup statically into the 'python'
> binary to make it work. It's ok for a short while, but I'd be very grateful
> if somebody pointed me to the source of the problem.
I'm not sure what is wrong here, but add is a method of the ufunc module,
so it sounds like that is where the problem is. The Numeric module
imports all of the methods of the ufunc module and so a NameError in this
circumstance seems to imply that the ufunc module is not completely
functional.
Try importing ufunc alone and looking at it's methods:
>>> import ufunc
>>> dir(ufunc)
You should see add there somewhere. If it is not there then it is
probably a shared library problem. Perhaps someone with more experience
on Solaris may have some insights.
Good luck,
Travis