From ccamp_ul at lycos.com Tue Jul 1 08:13:54 2003 From: ccamp_ul at lycos.com (Cameron Camp) Date: Tue, 01 Jul 2003 12:13:54 +0000 Subject: [SciPy-dev] Hey Message-ID: -------------- next part -------------- An HTML attachment was scrubbed... URL: From email.administrator at knightfrank.com Tue Jul 1 10:47:28 2003 From: email.administrator at knightfrank.com (email.administrator at knightfrank.com) Date: Tue, 1 Jul 2003 15:47:28 +0100 Subject: [SciPy-dev] Secure Messaging Non-Delivery Report: Re: Movie Message-ID: #################################################### When scipy-dev at scipy.net sent e-mail to WORCS at KNIGHTFRANK.COM (1/1/2) #################################################### Explanation: The message was blocked by TFS Secure Messaging ***************************************************** This e-mail is intended solely for its addressee. The contents are confidential and may be legally privileged. Any unauthorised access, disclosure, use or copying is strictly prohibited and may be unlawful. A list of Partners of Knight Frank can be inspected on our Internet Web Site at www.knightfrank.com under Knight Frank - Particulars of Ownership. ***************************************************** From rbojosi at netscape.net Tue Jul 1 10:49:39 2003 From: rbojosi at netscape.net (DR ROBERT BOJOSI) Date: Tue, 1 Jul 2003 16:49:39 +0200 Subject: [SciPy-dev] (no subject) Message-ID: <20030701160643.359B53EB09@www.scipy.com> DR.ROBERT BOJOSI The Zambia Home Office. The Hague. The Netherlands. Tel: +31 63 055 8460. Fax:+31 6 47 56 09 08 Email:robertbojosi at netscape.net Attn: Dear Sir REQUEST FOR ASSISTANCE IN A FINANCIAL TRANSACTION. I am interested in your partnership in business dealing. This business proposal I wish to intimate you with is of mutual benefit and it's success is entirely based on mutual trust, cooperation and a high level of confidentiality as regard this transaction. I am representing the board of the contract award and monitoring committee of the Zambian Ministry of Mining and Resources. I am seeking your assistance to enable me transfer the sum of US$15,500,000.00 (Fifteen Million, Five Hundred Thousand United States Dollars) into your private/company account. The fund came about as a result of a contract awarded and executed for and on behalf of my Ministry. The contract was supposed to be awarded to two foreign contractors to the tune of US$100,000,000.00 (One hundred Million United States Dollars). But in the course of negotiation, the contract was awarded to a Bulgarian contractor at the cost of US$115,500,000.00 (One hundred and Forty-nine Million, Five Hundred Thousand United States Dollars) to my benefit unknown to the contractor. This contract has been satisfactorily executed and inspected as the Bulgarian firm is presently securing payment from my Ministry, where our Board is in-charge of all foreign contract payment approval. As a civil servant still in active government service, I am forbidden by law to operate an account outside the shores of Zambia. Hence this message to you seeking your assistance so as to enable me present your private/company account details as a beneficiary of contractual claims alongside that of the Bulgarian contractor, to enable me transfer the difference of US $15,500,000.00 (Fifteen Million, Five Hundred Thousand United States Dollars) into your provided account. On actualization, the fund will be disbursed as stated below. 20% for you and 80% for us. What We request from you. 1,your names, company name if any- otherwise we will use your name to register the company. 2, your address, tel, fax and cell phone numbers. The name of your bank, address. Account number and swift code or routing number. The of account holder/beneficiary. All logistics are in place and all modalities worked out for a smooth actualization of the transaction within the next few working days of commencement. For further details as to the workability of this transaction, please reach me as soon as possible for further clarification. Please, call me on receipt of this mail on Tel: 0031-63 055 8460 Thank you while I await your urgent response. Yours Sincerely, DR. robert bojosi ROBERTBOJOSI at netscape.net IF YOU ARE INTERESTED TO CONCLUDE THIS WITH US-PLS FILL THIS FORM AND FAX BACK TO ME TO LEGALISE THIS TRANSACTION AT THE COURT.I WILL SEND YOU FURTHER DETAILS. REPUBLIC OF ZAMBIA . MINISTRY OF J USTICE LUSAKA. INDEMNITY APPLICATION FORM FOR TRANSFER OF RIGHTS & PRIVILEGES Date: Under Ministerial Law of Section MBR/D/E-00ZB You are required to fill this form and return promptly to this office for final confirmation and procurement : (A) Name of Applicant (Beneficiary): ................................................................................................................................... (B) Address: ............................................................................................................................................................................. (C) Phone: ............................................................... Fax: ....................................................................................................... (D) Banker: .............................................................................................................................................................................. (E) Banker's Address: ........................................................................................................................................................... (F) Phone: .................................................................. Fax: ..................................................................................................... (G) Account No.: .......................................................... Sort Code: ....................................................................................... (H) Contract No.: ..................................................................................................................................................................... (I) Amount words: ................................................................................................................................................................. (J) Your Signature ................................................................ CHECKING OFFICER: ................................................ AUTHORISING OFFICER............................................................... SIGN: ....................................................................... SIGN: .................................................................................................. DATE: ............................................................................. DATE................................................................................................. . APPROVED BY: ................................................................................................................................................ SIGN: ................................................................................................................................................................. DATE: .................................................................................................................................................................. From rosaliehayes_nm at australiandomainregistration.com Wed Jul 2 02:52:09 2003 From: rosaliehayes_nm at australiandomainregistration.com (Rosalie Hayes) Date: Wed, 02 Jul 2003 06:52:09 +0000 Subject: [SciPy-dev] Help Please.. Message-ID: -------------- next part -------------- An HTML attachment was scrubbed... URL: From Norman.Shelley at motorola.com Wed Jul 2 15:38:03 2003 From: Norman.Shelley at motorola.com (Norman Shelley) Date: Wed, 02 Jul 2003 12:38:03 -0700 Subject: [SciPy-dev] Future of other optimizers, e.g. dn2fb, ... Message-ID: <3F03349A.D9979933@motorola.com> I'm pleased to see so much in scipy. I noticed the minpack optimizer is wrapped and included. Are there any plans to wrap other optimizers like, http://www.netlib.org/port/dn2fb.f and others at that site and other sites? Just curious. Norman Shelley From nwagner at mecha.uni-stuttgart.de Fri Jul 4 04:03:18 2003 From: nwagner at mecha.uni-stuttgart.de (Nils Wagner) Date: Fri, 04 Jul 2003 10:03:18 +0200 Subject: [SciPy-dev] The Matrix Computation Toolbox Message-ID: <3F0534C6.DFC6F14F@mecha.uni-stuttgart.de> Hi all, There exists an interesting matrix computation toolbox that has been developed by Nicholas J. Higham. http://www.maths.man.ac.uk/~higham/mctoolbox/ The toolbox is distributed under the terms of the GNU General Public License (version 2 of the License, or any later version) as published by the Free Software Foundation. Maybe the features of scipy can benefit from this toolbox. However, it is a collection of Matlab m-files. Nils From dmorrill at austin.rr.com Mon Jul 7 12:19:53 2003 From: dmorrill at austin.rr.com (David C. Morrill) Date: Mon, 7 Jul 2003 11:19:53 -0500 Subject: [SciPy-dev] Updated version of Python 2.2.3 (Enthought Edition) now available Message-ID: <002701c344a3$99220f10$757ba8c0@brutus> Just wanted to let everyone know that an updated version of Python 2.2.3 (Enthought Edition) for Windows is available at: http://www.enthought.com/python/ The main changes are several new packages: a.. MayaVi 1.2: 3D Data Visualization Tool b.. ScientificPython 2.4.3: A collection of Python modules for scientific computing c.. F2PY 2.32.225-1419: A Fortran to Python interface generator d.. ZODB3 3.1: ZODB and ZEO Object DataBase e.. Gadfly 1.0.0: An SQL Relational Database in Python f.. PySQLite 0.4.3: A Python Extension for the SQLite embedded relational database Also, nearly all of the documentation has been collected together into a single, searchable Windows Compiled Help Module (.chm) file. Enjoy! The Python guys at Enthought, Inc. From 237995 at yahoo.com Tue Jul 8 04:11:40 2003 From: 237995 at yahoo.com (237995 at yahoo.com) Date: Tue, 08 Jul 2003 03:11:40 -0500 Subject: [SciPy-dev] 237995 Powerful Monitoring and Surveillance Software 1633239473 Message-ID: 79C05EE5-3AF73B3B-5B75C6C4-3CAB0761-1BD85D12@delphi.com An HTML attachment was scrubbed... URL: From Marketinger at yaomail.vicp.net Tue Jul 8 01:58:37 2003 From: Marketinger at yaomail.vicp.net (Judy) Date: Tue, 8 Jul 2003 13:58:37 +0800 Subject: [SciPy-dev] Promote Your Clients Message-ID: <20030708071636.84A403EB09@www.scipy.com> E-mail Marketing is one of the most effective and inexpensive ways to promote your products and services. We offer a complete E-mail Marketing solution with quality service and the lowest prices. The result is that you will enjoy more success. 1. Targeted E-mail Addresses We can supply targeted e-mail addresses according to your requirements, which are compiled only on your order, such as region / country / field / occupation / Domain Name etc. We will customize your customer e-mail addresses. * We have millions of e-mail addresses in a wide variety of categories. 2. Targeted Mailing If you are worried about any complications or consequences with sending out targeted e-mails, or want to avoid the work of sending out targeted e-mails. We will do it for you! We will send your e-mail message to your targeted customers. * We can Bullet-Proof your Web Site. We also offer a wide variety of e-mail marketing software. For more details, you can refer to our web site: http://www.lead-host.com Our services will help you get more business opportunities. Regards! Mrs Judy Customer Services www.lead-host.com E-mail Marketing, at Great Fees. ************************************************************************ Receiving this email because you registered to receive special offers from one of our partners. If you would prefer not to receive future e-mail, Click here: http://remove.contact.cc/ ************************************************************************ From mandrews at nomore-accents.com Tue Jul 8 10:11:42 2003 From: mandrews at nomore-accents.com (Martin Andrews) Date: Tue, 08 Jul 2003 09:11:42 -0500 Subject: [SciPy-dev] - Are you learning English? Message-ID: <20030708153354.1BAC13EB09@www.scipy.com> An HTML attachment was scrubbed... URL: From dwzba848t at influentialdelight.com Wed Jul 9 03:44:04 2003 From: dwzba848t at influentialdelight.com (Luisa England) Date: Wed, 09 Jul 03 07:44:04 GMT Subject: [SciPy-dev] Women are not illogical - learn why Message-ID: An HTML attachment was scrubbed... URL: From mandrews at no-moreaccent.com Wed Jul 9 06:25:50 2003 From: mandrews at no-moreaccent.com (Martin Andrews) Date: Wed, 09 Jul 2003 05:25:50 -0500 Subject: [SciPy-dev] - Are you learning English? Message-ID: <20030709114752.D2F073EB09@www.scipy.com> An HTML attachment was scrubbed... URL: From mandrews at no-moreaccent.com Wed Jul 9 13:47:26 2003 From: mandrews at no-moreaccent.com (Martin Andrews) Date: Wed, 09 Jul 2003 12:47:26 -0500 Subject: [SciPy-dev] - Are you learning English? Message-ID: <20030709190958.06C8E3EB09@www.scipy.com> An HTML attachment was scrubbed... URL: From p.berkes at biologie.hu-berlin.de Thu Jul 10 08:23:22 2003 From: p.berkes at biologie.hu-berlin.de (Pietro Berkes) Date: Thu, 10 Jul 2003 14:23:22 +0200 Subject: [SciPy-dev] matrixmultiply alias Message-ID: <200307101423.22481.p.berkes@biologie.hu-berlin.de> Hi! I would like to propose a function that could be used as alias for scipy.matrixmultiply (code attached in 'matrixmultiply.py'). The function delegate the work to scipy.dot if the arguments it receives aren't 2D, otherwise it uses the BLAS routine 'gemm'. If BLAS fails, an appropriate error message is raised. The function should now be backward-compatible and is sensibly faster than the original one for matrices larger than 40x40. Here are the benchmarks (the code to perform the benchmarks is included in 'matmult_bench.py'): Multiplying matrices of type f ================================== | contiguous | non-contiguous ---------------------------------------------- size | scipy | matmult | scipy | matmult 20 | 0.11 | 0.24 | 0.11 | 0.24 (secs for 2000 calls) 50 | 0.99 | 0.60 | 1.01 | 0.60 (secs for 2000 calls) 75 | 3.10 | 1.44 | 3.09 | 1.46 (secs for 2000 calls) 100 | 3.64 | 0.80 | 3.62 | 0.83 (secs for 1000 calls) 500 | 5.06 | 0.25 | 5.08 | 0.26 (secs for 4 calls) 1000 | 28.05 | 0.94 | 27.96 | 0.94 (secs for 2 calls) Multiplying matrices of type d ================================== | contiguous | non-contiguous ---------------------------------------------- size | scipy | matmult | scipy | matmult 20 | 0.10 | 0.26 | 0.10 | 0.26 (secs for 2000 calls) 50 | 1.01 | 0.68 | 1.02 | 0.69 (secs for 2000 calls) 75 | 3.44 | 1.87 | 3.48 | 1.91 (secs for 2000 calls) 100 | 3.59 | 1.55 | 3.59 | 1.49 (secs for 1000 calls) 500 | 8.02 | 0.50 | 8.01 | 0.52 (secs for 4 calls) 1000 | 36.31 | 1.86 | 36.15 | 1.83 (secs for 2 calls) Regards, Pietro. -------------- next part -------------- A non-text attachment was scrubbed... Name: matrixmultiply.py Type: text/x-python Size: 2407 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: matmult_bench.py Type: text/x-python Size: 4758 bytes Desc: not available URL: From mandrews at nomore-accents.com Fri Jul 11 14:21:17 2003 From: mandrews at nomore-accents.com (Martin Andrews) Date: Fri, 11 Jul 2003 13:21:17 -0500 Subject: [SciPy-dev] - Are you learning ESL or TESL? Message-ID: <20030711194437.1D0503EB09@www.scipy.com> Dear Reader, No More Accent is an accent reduction product, which helps to remove accents. For example, an Asian lady (or any nationality) may have communication problems with her English when speaking to her work mates or friends. With that in mind, No More Accent teaches Accent Reduction methodologies to help in this area. In fact, No More Accent can be used in almost any situation where an accent reduction is required. Even Actors within the Entertainment Industry have turned to No More Accent as a way to lose or gain an accent of another nationality. No More Accent is available in two versions, CD-ROM and VHS Cassette Tape (PAL and NTSC formats) both can be purchased directly via the www.nomore-accents.com website. If you would like further information about our products or services, please contact myself when you have a moment. Best Wishes... Martin Andrews Marketing Co-ordinator http://www.nomore-accents.com Your #1 Resource for Accent Reduction! From TBjaded at aol.com Sat Jul 12 13:15:08 2003 From: TBjaded at aol.com (TBjaded at aol.com) Date: Sat, 12 Jul 2003 13:15:08 EDT Subject: [SciPy-dev] you suck Message-ID: <12d.2d68c252.2c419c1c@aol.com> you filthy swine, i hope you burn in hell!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! -------------- next part -------------- An HTML attachment was scrubbed... URL: From j_r_fonseca at yahoo.co.uk Mon Jul 14 18:57:37 2003 From: j_r_fonseca at yahoo.co.uk (=?iso-8859-15?Q?Jos=E9?= Fonseca) Date: Mon, 14 Jul 2003 23:57:37 +0100 Subject: [SciPy-dev] Debian packages of scipy and f2py Message-ID: <20030714225737.GB1869@prometeu> I've make Debian packages of scipy and f2py. They are available from my apt-get repository at: http://jrfonseca.dyndns.org/debian/ I referred to Joe Reinhardt's packages http://people.debian.org/~jmr/ but have done these packages from scratch. These come for all python versions in debian unstable (2.1, 2.2 and 2.3). I've run the standard tests successfully but I'd appreciate that problems with these packages to be reported to me [privately] as I'd like to have these on Debian proper at some time. One tricky issue I'd like to address with the scipy and f2py maintainers is the the scipy_distutils. Both scipy and f2py attempt to install to /usr/lib/python*/site-packages/scipy_distutils and at least on Debian this is not admissible so the packages conflict with each other and are mutually exclusive. This is particularly even more annoying for me, since for building scipy f2py is needed, so I need to install and uninstall them as a maniac. Is there any solution that allow both these packages to happily co-exist in a tree without overlapping? Enjoy, Jos? Fonseca PS: Please CC'me on any reply as I'm not subscribed to all these lists. From nwagner at mecha.uni-stuttgart.de Tue Jul 15 03:00:34 2003 From: nwagner at mecha.uni-stuttgart.de (Nils Wagner) Date: Tue, 15 Jul 2003 09:00:34 +0200 Subject: [SciPy-dev] [Fwd: [SciPy-user] Possible bug in linalg.lu] Message-ID: <3F13A692.ABEDB2D2@mecha.uni-stuttgart.de> Hi all, It would be great, if linalg.lu handles complex matrices as well. Thanks in advance Nils -------------- next part -------------- An embedded message was scrubbed... From: Pietro Berkes Subject: Re: [SciPy-user] Possible bug in linalg.lu Date: Mon, 14 Jul 2003 16:23:15 +0200 Size: 3480 URL: From Ralf_Ahlbrink at web.de Wed Jul 16 17:50:14 2003 From: Ralf_Ahlbrink at web.de (Ralf Ahlbrink) Date: Wed, 16 Jul 2003 23:50:14 +0200 Subject: [SciPy-dev] Debian packages of scipy and f2py In-Reply-To: <20030714225737.GB1869@prometeu> References: <20030714225737.GB1869@prometeu> Message-ID: <200307162350.15021.Ralf_Ahlbrink@web.de> Am Dienstag 15 Juli 2003 00:57 schrieb Jos? Fonseca: > I've make Debian packages of scipy and f2py. They are available from my > apt-get repository at: > > http://jrfonseca.dyndns.org/debian/ Hello Jos?! Please have a look at http://www.scipy.org/Members/RalfA/packages/Debian-woody-DEB/ for package related files (you have to have an account on scipy.org) > > I referred to Joe Reinhardt's packages http://people.debian.org/~jmr/ > but have done these packages from scratch. These come for all python > versions in debian unstable (2.1, 2.2 and 2.3). dito > > I've run the standard tests successfully but I'd appreciate that > problems with these packages to be reported to me [privately] as I'd > like to have these on Debian proper at some time. lintian was not really happy with my packages ... I would be happy if there will be a proper package in Debian. > > One tricky issue I'd like to address with the scipy and f2py maintainers > is the the scipy_distutils. Both scipy and f2py attempt to install to > /usr/lib/python*/site-packages/scipy_distutils and at least on Debian > this is not admissible so the packages conflict with each other and are > mutually exclusive. This is particularly even more annoying for me, > since for building scipy f2py is needed, so I need to install and > uninstall them as a maniac. Is there any solution that allow both these > packages to happily co-exist in a tree without overlapping? Replaces & Provides in debian/control - have a look at the above mentioned packages > > > Enjoy, > > Jos? Fonseca > > > PS: Please CC'me on any reply as I'm not subscribed to all these lists. > Ralf. From freedryk at ocean.washington.edu Mon Jul 21 18:09:39 2003 From: freedryk at ocean.washington.edu (Jordan Dawe) Date: Mon, 21 Jul 2003 15:09:39 -0700 Subject: [SciPy-dev] gcc 3.2.3 and weave Message-ID: <3F1C64A3.9050108@ocean.washington.edu> I believe that weave is currently broken with gcc-3.2.3--something to do with a change in the definitions of C++ headers. I say this because a) while installing fine, weave's test suite conks out with massive errors, all gcc-related, b) The netcdf library has a similar problem compiling its C++ componets under gcc-3.2.3 and c) the netcdf developers say that it's a problem with changes to the C++ headers in gcc-3.2.3, and have fixed em. Not sure if this is the right place to report this, just thought you might want to know. Jordan From cookedm at physics.mcmaster.ca Mon Jul 21 20:12:50 2003 From: cookedm at physics.mcmaster.ca (David M. Cooke) Date: Mon, 21 Jul 2003 20:12:50 -0400 Subject: [SciPy-dev] Random fixes Message-ID: <20030722001250.GA26509@arbutus.physics.mcmaster.ca> Just some stuff I found running pychecker over a few files... -- |>|\/|< /--------------------------------------------------------------------------\ |David M. Cooke http://arbutus.physics.mcmaster.ca/dmc/ |cookedm at physics.mcmaster.ca -------------- next part -------------- Index: Lib/integrate/ode.py =================================================================== RCS file: /home/cvsroot/world/scipy/Lib/integrate/ode.py,v retrieving revision 1.1 diff -c -u -r1.1 ode.py --- Lib/integrate/ode.py 3 Feb 2002 19:08:07 -0000 1.1 +++ Lib/integrate/ode.py 22 Jul 2003 01:36:09 -0000 @@ -200,6 +200,7 @@ if re.match(name,cl.__name__,re.I): print 'Found integrator',cl.__name__ return cl + return None class IntegratorBase: Index: Lib/linalg/interface_gen.py =================================================================== RCS file: /home/cvsroot/world/scipy/Lib/linalg/interface_gen.py,v retrieving revision 1.14 diff -c -u -r1.14 interface_gen.py --- Lib/linalg/interface_gen.py 13 Oct 2002 23:09:28 -0000 1.14 +++ Lib/linalg/interface_gen.py 22 Jul 2003 01:36:09 -0000 @@ -1,7 +1,7 @@ #!/usr/bin/env python import string,os -import pre as re +import re def all_subroutines(interface_in): # remove comments Index: Lib/optimize/optimize.py =================================================================== RCS file: /home/cvsroot/world/scipy/Lib/optimize/optimize.py,v retrieving revision 1.36 diff -c -u -r1.36 optimize.py --- Lib/optimize/optimize.py 17 Sep 2002 16:13:22 -0000 1.36 +++ Lib/optimize/optimize.py 22 Jul 2003 01:36:09 -0000 @@ -590,7 +590,7 @@ while Num.add.reduce(abs(ri)) > termcond: if fhess is None: if fhess_p is None: - Ap = apply(approx_fhess_p,(xk,psupi,fprime,espilon)+args) + Ap = apply(approx_fhess_p,(xk,psupi,fprime,epsilon)+args) gcalls = gcalls + 2 else: Ap = apply(fhess_p,(xk,psupi)+args) @@ -782,7 +782,7 @@ flag = 1 fval = fx if disp > 0: - _endprint(x, flag, fval, maxfun, tol, disp) + _endprint(x, flag, fval, maxfun, xtol, disp) if full_output: return xf, fval, flag, num else: @@ -826,7 +826,7 @@ assert ((fb 1000: - raise RunTimeError, "Too many iterations." + raise RuntimeError, "Too many iterations." if (w-xc)*(xb-w) > 0.0: fw = apply(func, (w,)+args) funcalls += 1 @@ -1018,7 +1018,7 @@ return xa, xb, xc, fa, fb, fc, funcalls -global _powell_funcalls +_powell_funcalls = 0 def _myfunc(alpha, func, x0, direc, args=()): funcargs = (x0 + alpha * direc,)+args Index: Lib/weave/build_tools.py =================================================================== RCS file: /home/cvsroot/world/scipy/Lib/weave/build_tools.py,v retrieving revision 1.25 diff -c -u -r1.25 build_tools.py --- Lib/weave/build_tools.py 29 Mar 2003 06:24:37 -0000 1.25 +++ Lib/weave/build_tools.py 22 Jul 2003 01:36:09 -0000 @@ -518,7 +518,7 @@ libraries, library_dirs, runtime_library_dirs, - None, # export_symbols, we do this in our def-file + _, # export_symbols, we do this in our def-file debug, extra_preargs, extra_postargs, From travis at enthought.com Thu Jul 24 21:53:20 2003 From: travis at enthought.com (Travis N. Vaught) Date: Thu, 24 Jul 2003 20:53:20 -0500 Subject: [SciPy-dev] [ANN] SciPy 03 Workshop - Scientific Computing with Python Message-ID: <001301c3524f$89b943e0$0200a8c0@tvlaptop> ------------------ SciPy '03 Workshop ------------------ The SciPy '03 Workshop on Scientific Computing with Python is being held at Cal Tech again this year. Online registration is now available. After last year's success, we expect another great turnout with compelling presentations/tutorials. It is a two day workshop held September 11-12 in Pasadena, CA. More information may be found at: http://www.scipy.org/site_content/scipy03 Online registration is available at: https://www.enthought.com/scipy03 Please register early. Last minute registrants will have to suffer the indignity of a handwritten name tag. It currently looks like breakfast (bagels, etc.), lunch and supper will be provided Thursday, and breakfast and lunch will be provided Friday. Discussion about the workshop may be directed to the SciPy-user mailing list (http://www.scipy.org/site_content/MailList) to which you may post messages at scipy-user at scipy.org. Presenters: If you would like to present at the workshop, there are two formats: - Lightning Talks: Short (5-15 minute) presentations of a particular scientific problem solved with Python. - Presentations: Traditional presentation format ~45 minute talks. If you would like to present, contact Eric Jones at eric at enthought.com. For any other questions, please feel free to call or email Travis Vaught: travis at enthought.com, (512)536-1057. You can call the same number to register via phone (Business Hours CST). Links: SciPy.org community site: http://www.scipy.org Workshop Home: http://www.scipy.org/site_content/scipy03 Summary of last year's workshop: http://www.scipy.org/site_content/scipy02/scipy02summary.htm From a.schmolck at gmx.net Fri Jul 25 13:36:15 2003 From: a.schmolck at gmx.net (Alexander Schmolck) Date: 25 Jul 2003 18:36:15 +0100 Subject: [SciPy-dev] matrixmultiply alias In-Reply-To: <200307101423.22481.p.berkes@biologie.hu-berlin.de> References: <200307101423.22481.p.berkes@biologie.hu-berlin.de> Message-ID: Pietro Berkes writes: > Hi! > > I would like to propose a function that could be used as alias for > scipy.matrixmultiply (code attached in 'matrixmultiply.py'). The function > delegate the work to scipy.dot if the arguments it receives aren't 2D, > otherwise it uses the BLAS routine 'gemm'. If BLAS fails, an appropriate > error message is raised. > The function should now be backward-compatible and is sensibly faster than the > original one for matrices larger than 40x40. Here are the benchmarks (the > code to perform the benchmarks is included in 'matmult_bench.py'): For those who are interested, a similiar patch ("dotblas" submitted by me but coded mainly by Richard Everson) is now part of Numeric (version 23). For 1000x1000 single float matrices I think the speedup is around factor 80. 'as