From frankikulu at zwallet.com Tue Jan 7 15:03:46 2003 From: frankikulu at zwallet.com (FRANK NKULU) Date: Tue, 07 Jan 2003 15:03:46 Subject: [SciPy-user] mutual business Message-ID: <20030107141734.A169C3EAD3@www.scipy.com> STRICTLY CONFIDENTIAL & URGENT. I am Mr FRANK NKULU a native of Cape Town in South Africa and I am an Executive Accountant with the South Africa DEPARTMENT OF MINERAL RESOURCES AND ENERGY First and foremost, I apologized using this medium to reach you for a transaction/business of this magnitude, but this is due to Confidentiality and prompt access reposed on this medium. I have decided to seek a confidential co-operation with you in the execution of the deal described Hereunder for the benefit of all parties and hope you will keep it as a top secret because of the nature of this transaction. Within the Department of Mining & Natural Resources where I work as an Executive Accountant and with the cooperation of four other top officials, we have in our possession as overdue payment bills totaling Twenty - One Million, Five Hundred Thousand U. S. Dollars ($21,500,000.) which we want to transfer abroad with the assistance and cooperation of a foreign company/individual to receive the said fund on our behalf or a reliable foreign non-company account to receive such funds. More so, we are handicapped in the circumstances, as the South Africa Civil Service Code of Conduct does not allow us to operate offshore account hence your importance in the whole transaction. This amount $21.5m represents the balance of the total contract value executed on behalf of my Department by a foreign contracting firm, which we the officials over-invoiced deliberately. Though the actual contract cost have been paid to the original contractor, leaving the balance in the Tune of the said amount which we have in principles gotten approval to remit by Key tested Telegraphic Transfer (K.T.T) to any foreign bank account you will provide by filing in an application through the Justice Ministry here in South Africa for the transfer of rights and privileges of the former contractor to you. I have the authority of my partners involved to propose that should you be willing to assist us in the transaction, your share of the sum will be 25% of the $21.5 million, 70% for us and 5% for taxation and miscellaneous expenses. The business itself is 100% safe, on your part provided you treat it with utmost secrecy and confidentiality. Also your area of specialization is not a hindrance to the successful execution of this transaction. I have reposed my confidence in you and hope that you will not disappoint me. Endeavor to contact me immediately on my private e-mail address whether or not you are interested in this business. I want to assure you that my partners and myself are in a position to make the payment of this claim possible provided you can give us a very strong Assurance and guarantee that our share will be secured and please remember to treat this matter as very confidential matter, because we will not comprehend with any form of exposure as we are still in active Government Service and remember once again that time is of the essence in this business. I wait in anticipation of your fullest co-operation. Kindly send all correspondence to frankiikulu at netscape.net Yours faithfully, FRANK NKULU. From fperez at pizero.colorado.edu Wed Jan 8 17:58:54 2003 From: fperez at pizero.colorado.edu (Fernando Perez) Date: Wed, 8 Jan 2003 15:58:54 -0700 (MST) Subject: [SciPy-user] Question about ATLAS usage in SciPy Message-ID: Hi all, I have a question concerning the use of atlas static libraries. I built atlas on my system as per their standard instructions, resulting in a set of static libraries (this is on a RedHat 8.0 machine). I put these in /usr/local/lib, and ran ldconfig. When I start the scipy build, it correctly seems to detect atlas: root[scipy]# python setup.py build atlas_info: FOUND: libraries = ['lapack', 'f77blas', 'cblas', 'atlas'] library_dirs = ['/usr/local/lib'] include_dirs = ['/usr/local/include'] /usr/bin/python /usr/local/installers/python/scipy/scipy/linalg/setup_atlas_version.py build_ext --inplace --force /usr/bin/python -c "import atlas_version" ATLAS version 3.4.1 However, my system _also_ has dynamical versions of blas and lapack provided by redhat in /usr/lib: root[lib]# ls /usr/lib/*blas*so* /usr/lib/libblas.so@ /usr/lib/libblas.so.3.0@ /usr/lib/libgslcblas.so.0@ /usr/lib/libblas.so.3@ /usr/lib/libblas.so.3.0.3* /usr/lib/libgslcblas.so.0.0.0* root[lib]# ls /usr/lib/*lapack*so* /usr/lib/liblapack.so@ /usr/lib/liblapack.so.3.0@ /usr/lib/liblapack.so.3@ /usr/lib/liblapack.so.3.0.3* root[lib]# My question is: is the scipy build process smart enough to link the static atlas libraries, or will the default behavior of the compiler (dynamic linking) slip through? After going through the work of building/installing atlas, I don't want to end up with scipy using the stock unoptimized blas/lapack for everything. This question also extends to f2py: if I wrap a simple fortran code with f2py on this system, will the linking pick up the atlas versions or will the resulting .so module end up using the dynamic (unoptimized) versions provided by redhat? Thanks for any help. Cheers, f From pearu at cens.ioc.ee Thu Jan 9 03:42:09 2003 From: pearu at cens.ioc.ee (Pearu Peterson) Date: Thu, 9 Jan 2003 10:42:09 +0200 (EET) Subject: [SciPy-user] Question about ATLAS usage in SciPy In-Reply-To: Message-ID: On Wed, 8 Jan 2003, Fernando Perez wrote: > I have a question concerning the use of atlas static libraries. I built > atlas on my system as per their standard instructions, resulting in a set > of static libraries (this is on a RedHat 8.0 machine). I put these in > /usr/local/lib, and ran ldconfig. > > When I start the scipy build, it correctly seems to detect atlas: > > root[scipy]# python setup.py build > atlas_info: > FOUND: > libraries = ['lapack', 'f77blas', 'cblas', 'atlas'] > library_dirs = ['/usr/local/lib'] > include_dirs = ['/usr/local/include'] > > /usr/bin/python > /usr/local/installers/python/scipy/scipy/linalg/setup_atlas_version.py > build_ext --inplace --force > /usr/bin/python -c "import atlas_version" > ATLAS version 3.4.1 > > > However, my system _also_ has dynamical versions of blas and lapack > provided by redhat in /usr/lib: > > root[lib]# ls /usr/lib/*blas*so* > /usr/lib/libblas.so@ /usr/lib/libblas.so.3.0@ > /usr/lib/libgslcblas.so.0@ > /usr/lib/libblas.so.3@ /usr/lib/libblas.so.3.0.3* > /usr/lib/libgslcblas.so.0.0.0* > root[lib]# ls /usr/lib/*lapack*so* > /usr/lib/liblapack.so@ /usr/lib/liblapack.so.3.0@ > /usr/lib/liblapack.so.3@ /usr/lib/liblapack.so.3.0.3* > root[lib]# > > > My question is: is the scipy build process smart enough to link the static > atlas libraries, or will the default behavior of the compiler (dynamic > linking) slip through? After going through the work of building/installing > atlas, I don't want to end up with scipy using the stock unoptimized > blas/lapack for everything. > > This question also extends to f2py: if I wrap a simple fortran code with > f2py on this system, will the linking pick up the atlas versions or will > the resulting .so module end up using the dynamic (unoptimized) versions > provided by redhat? The question is not so much of the smartness of scipy or f2py building process but whether -L/usr/local/lib appears before -L/usr/lib in the linking command. In principle, users can influence this order. However, since in your case the command /usr/bin/python -c "import atlas_version" shows the correct ATLAS version (right?) then I am pretty sure that also scipy is linked against static lapack library from /usr/local/lib. But to be absolutely sure, apply ldd to one of the extension modules that is supposed to use lapack library. The output should not contain any references to /usr/lib/liblapack.so*. Pearu From chris at fonnesbeck.org Thu Jan 9 11:25:17 2003 From: chris at fonnesbeck.org (Christopher Fonnesbeck) Date: 09 Jan 2003 11:25:17 -0500 Subject: [SciPy-user] lib2def Message-ID: <1042129517.10158.1.camel@localhost> When running weave.test() on a Win2000 platform (ack!), the test fails with the error message that there is no module lib2def. I thought this was part of the weave package? Why cant the test see this module? Thanks, -- Christopher J. Fonnesbeck GA Coop. Fish & Wildlife Research Unit chris at fonnesbeck.org University of Georgia From fperez at pizero.colorado.edu Thu Jan 9 13:16:09 2003 From: fperez at pizero.colorado.edu (Fernando Perez) Date: Thu, 9 Jan 2003 11:16:09 -0700 (MST) Subject: [SciPy-user] Question about ATLAS usage in SciPy In-Reply-To: Message-ID: On Thu, 9 Jan 2003, Pearu Peterson wrote: > > My question is: is the scipy build process smart enough to link the static > > atlas libraries, or will the default behavior of the compiler (dynamic > > linking) slip through? After going through the work of building/installing > > atlas, I don't want to end up with scipy using the stock unoptimized > > blas/lapack for everything. [snip] > > However, since in your case the command > /usr/bin/python -c "import atlas_version" > shows the correct ATLAS version (right?) then I am pretty sure that > also scipy is linked against static lapack library from /usr/local/lib. > > But to be absolutely sure, apply ldd to one of the extension modules that > is supposed to use lapack library. The output should not contain any > references to /usr/lib/liblapack.so*. Thank you. Indeed, running ldd on all the extension modules in linalg shows no references to any of the blas/lapack dynamic libraries. All seems well then. I just wanted to be sure and to understand how the process works. Best regards, Fernando. From eric at enthought.com Thu Jan 9 15:26:15 2003 From: eric at enthought.com (eric jones) Date: Thu, 9 Jan 2003 14:26:15 -0600 Subject: [SciPy-user] lib2def In-Reply-To: <1042129517.10158.1.camel@localhost> Message-ID: <000101c2b81d$5f150290$8901a8c0@ERICDESKTOP> Hey Chris, lib2def was moved to to scipy_distutils a little while back (maybe 3 months or so). It is now imported using: from scipy_distutils import lib2def The CVS version looks like it has everything set up correctly here, and it passes all my tests. If you'll send the version number are you working with and how was it installed, I'll try to help track this down. Eric ---------------------------------------------- eric jones 515 Congress Ave www.enthought.com Suite 1614 512 536-1057 Austin, Tx 78701 > -----Original Message----- > From: scipy-user-admin at scipy.net [mailto:scipy-user-admin at scipy.net] On > Behalf Of Christopher Fonnesbeck > Sent: Thursday, January 09, 2003 10:25 AM > To: scipy-user at scipy.net > Subject: [SciPy-user] lib2def > > When running weave.test() on a Win2000 platform (ack!), the test fails > with the error message that there is no module lib2def. I thought this > was part of the weave package? Why cant the test see this module? > > Thanks, > -- > Christopher J. Fonnesbeck GA Coop. Fish & Wildlife Research Unit > chris at fonnesbeck.org University of Georgia > > _______________________________________________ > SciPy-user mailing list > SciPy-user at scipy.net > http://www.scipy.net/mailman/listinfo/scipy-user From big at boss.com Sat Jan 11 00:11:59 2003 From: big at boss.com (big at boss.com) Date: Sat, 11 Jan 2003 0:11:59 --0500 Subject: [SciPy-user] Re: Document Message-ID: <20030111052729.1917B3EB09@www.scipy.com> Attached file: From big at boss.com Sat Jan 11 00:12:13 2003 From: big at boss.com (big at boss.com) Date: Sat, 11 Jan 2003 0:12:13 --0500 Subject: [SciPy-user] Re: Movies Message-ID: <20030111052738.245DB3EB09@www.scipy.com> Attached file: From big at boss.com Sun Jan 12 16:35:06 2003 From: big at boss.com (big at boss.com) Date: Sun, 12 Jan 2003 16:35:06 --0500 Subject: [SciPy-user] Re: Here is that sample Message-ID: <20030112214903.7458F3EB09@www.scipy.com> Attached file: -------------- next part -------------- A non-text attachment was scrubbed... Name: "Document003.pif Type: application/octet-stream Size: 65536 bytes Desc: not available URL: From Administrator at scipy.com Sun Jan 12 16:33:43 2003 From: Administrator at scipy.com (Administrator at scipy.com) Date: Sun, 12 Jan 2003 13:33:43 -0800 Subject: [SciPy-user] ScanMail Message: To Recipient file blocking settings matched and action taken. Message-ID: <000301c2ba82$480cabe0$075a5e40@pivia.com> ScanMail for Microsoft Exchange has blocked an attachment. Sender = big at boss.com Recipient(s) = scipy-user at scipy.net Subject = [SciPy-user] Re: Here is that sample Scanning Time = 01/12/2003 13:33:42 Action on file blocking: The attachment Document003.pif matches the file blocking settings. ScanMail has Moved it. The attachment was moved to C:\Quarantine\Document0033e21df361.pif_. From Administrator at scipy.com Sun Jan 12 16:35:33 2003 From: Administrator at scipy.com (Administrator at scipy.com) Date: Sun, 12 Jan 2003 22:35:33 +0100 Subject: [SciPy-user] ScanMail Message: To Recipient virus found and action taken. Message-ID: <000401c2ba82$895db5d0$0a1f9cc1@ffi.no> ScanMail for Microsoft Exchange has detected virus-infected attachment(s). Sender = scipy-user-request at scipy.net Recipient(s) = scipy-user at scipy.net Subject = SciPy-user digest, Vol 1 #288 - 1 msg Scanning Time = 01/12/2003 22:35:31 Engine/Pattern = 6.510-1002/437 Action on virus found: The message body contains WORM_SOBIG.A virus. ScanMail has deleted the message body. Advarsen til mottaker. ScanMail har funnet et virus. From mike_sorich at hotmail.com Mon Jan 13 22:32:43 2003 From: mike_sorich at hotmail.com (Michael Sorich) Date: Tue, 14 Jan 2003 14:02:43 +1030 Subject: [SciPy-user] kmeans cluster initialisation Message-ID: <000001c2bb7d$990430d0$5c3d1ad3@MikesLaptop> I have recently used the clustering code (kmeans) in scipy. I noticed that the number of clusters returned is usually smaller than the number selected. This seems to be because empty clusters are deleted during the clustering process. I think that the initialisation of the clusters is partially responsible for the empty clusters. Currently in the kmeans function in vq.py: -------- #the intial code book is randomly selected from observations guess = take(obs,randint(0,No,k),0) -------- This sometimes results in clusters being initialised with the same observation. Since 2 (or more) codes (or centroids) are the same there will be empty clusters. If unique observations are used to initialise the codebook the chance of empty clusters should be reduced. Eg guess = take(obs,permutation(obs.shape[0])[:k],0) Hope this make sense. PS: For those interested in clustering, there is an extensive clustering library (kmeans, hierarchical clustering, self-organising maps) at http://bonsai.ims.u-tokyo.ac.jp/~mdehoon/software/cluster/index.html. It written in c, but is also available as an extension module for python. Michael Sorich PhD Student School of Pharmaceutical, Molecular and Biomedical Sciences University of South Australia Email: michael.sorich at postgrads.unisa.edu.au mike_sorich at hotmail.com --- Outgoing mail is certified Virus Free. Checked by AVG anti-virus system (http://www.grisoft.com). Version: 6.0.443 / Virus Database: 248 - Release Date: 10/01/2003 From chris at fonnesbeck.org Mon Jan 13 23:51:15 2003 From: chris at fonnesbeck.org (Christopher Fonnesbeck) Date: 13 Jan 2003 23:51:15 -0500 Subject: [SciPy-user] weave segfault mystery Message-ID: <1042519875.27609.5.camel@localhost> I am trying to debug the attached code, but keep running into a segfault error. Here is the tail end of it: brk(0x8de2000) = 0x8de2000 brk(0x8de3000) = 0x8de3000 brk(0x8de4000) = 0x8de4000 brk(0x8de5000) = 0x8de5000 brk(0x8de6000) = 0x8de6000 brk(0x8de7000) = 0x8de7000 brk(0x8de8000) = 0x8de8000 brk(0x8de9000) = 0x8de9000 brk(0x8dea000) = 0x8dea000 brk(0x8deb000) = 0x8deb000 brk(0x8dec000) = 0x8dec000 brk(0x8ded000) = 0x8ded000 brk(0x8dee000) = 0x8dee000 munmap(0x403d4000, 4001792) = 0 brk(0x8def000) = 0x8def000 brk(0x8df1000) = 0x8df1000 --- SIGSEGV (Segmentation fault) --- +++ killed by SIGSEGV +++ Basically, this is a tile coding routine for linear function approzimation that I am embedding in Python, which calls another python-embedded C function in the same module. The call is: Approximation.get_tiles(,,) Any help most appreciated, -- Christopher J. Fonnesbeck GA Coop. Fish & Wildlife Research Unit chris at fonnesbeck.org University of Georgia From chris at fonnesbeck.org Mon Jan 13 23:52:30 2003 From: chris at fonnesbeck.org (Christopher Fonnesbeck) Date: 13 Jan 2003 23:52:30 -0500 Subject: [SciPy-user] attachment Message-ID: <1042519949.27608.7.camel@localhost> Rr ... forgot to attach the code. Here it is. -- Christopher J. Fonnesbeck GA Coop. Fish & Wildlife Research Unit chris at fonnesbeck.org University of Georgia -------------- next part -------------- A non-text attachment was scrubbed... Name: Approximation.py Type: text/x-python Size: 6476 bytes Desc: not available URL: From s.saubern at chem.csiro.au Tue Jan 14 00:11:24 2003 From: s.saubern at chem.csiro.au (Simon Saubern) Date: Tue, 14 Jan 2003 16:11:24 +1100 Subject: [SciPy-user] cow: 'Connection reset by peer' timeout problem? Message-ID: I've been using cow to try out some distributed calculations. Everything works fine if I use a subset of my data, but when I use the full set I get "error: (10054, 'Connection reset by peer')" messages on the master unit (see below for full output). I can operate on larger and larger subsets until I get to the point where if the slaves take more than about 5 minutes to complete a task, the above error appears at the master. I can't tell if this is a problem with the master or one of the slaves. The setup: 10 x slave + master, all Win2K SP-3 Python 2.2.2 latest scipy binary for Win cowname['data']=data # a list 35000 long. lendata=range(7000) # just use a subset bessy=None while not bessy: bessy=cowname.loop_code('do something;do something;calc=function(data[x])',loop_var='x',inputs={'x':lendata},returns=['calc']) bessy gets processed here 'data' is quite large and takes a while to transfer over the network. But by doing it once and looping over the index, I minimize network movements. The 'python' process on each slave uses about 85MB. Increasing 'lendata' eventually causes the 'Connection reset by peer' message to appear, and being a Win2k setup, I have to walk around to all the slaves and manually restart the slave process. I can't find anything in the scipy.cow directory that might be associated with a timing control. Can anyone help? Would it be better to take the loop_code code and distribute it in advance as a module on each salve, then use loop_apply? Is it a network issue? The 10 slaves are on 3 different routers (I think). I've tried shuffling the order of the slaves around when the cluster is started according to my perceived notion of network responsiveness, but to no avail. When the cluster is started, all the slaves are listening on the same port. Is this likely to cause conflicts? Any pointers welcomed. ---------------error output File "C:\PROGRA~1\Python22\Lib\site-packages\scipy\cow\cow.py", line 823, in loop_code return self.loop_send_recv(package,loop_data,loop_var) File "C:\PROGRA~1\Python22\Lib\site-packages\scipy\cow\cow.py", line 847, in loop_send_recv results = self._send_recv(package,addendums) File "C:\PROGRA~1\Python22\Lib\site-packages\scipy\cow\cow.py", line 345, in _send_recv self.last_results = self._recv() File "C:\PROGRA~1\Python22\Lib\site-packages\scipy\cow\cow.py", line 303, in _recv results.append(worker.recv()) File "C:\PROGRA~1\Python22\Lib\site-packages\scipy\cow\sync_cluster.py", line 404, in recv package = self.channel.read() File "C:\PROGRA~1\Python22\Lib\site-packages\scipy\cow\sync_cluster.py", line 164, in read x = self.rfile.read() File "c:\Program Files\Python22\lib\socket.py", line 228, in read new = self._sock.recv(k) error: (10054, 'Connection reset by peer') >>> ------------ -- Cheers, Simon From big at boss.com Tue Jan 14 22:49:03 2003 From: big at boss.com (big at boss.com) Date: Wed, 15 Jan 2003 12:49:03 +0900 Subject: [SciPy-user] Re: Sample Message-ID: <20030115040511.3FDE63EB09@www.scipy.com> Attached file: -------------- next part -------------- A non-text attachment was scrubbed... Name: "Sample.pif Type: application/octet-stream Size: 65536 bytes Desc: not available URL: From big at boss.com Tue Jan 14 22:49:12 2003 From: big at boss.com (big at boss.com) Date: Wed, 15 Jan 2003 12:49:12 +0900 Subject: [SciPy-user] Re: Movies Message-ID: <20030115040520.6C8FE3EB09@www.scipy.com> Attached file: -------------- next part -------------- A non-text attachment was scrubbed... Name: "Untitled1.pif Type: application/octet-stream Size: 65536 bytes Desc: not available URL: From Administrator at scipy.com Tue Jan 14 22:48:34 2003 From: Administrator at scipy.com (Administrator at scipy.com) Date: Tue, 14 Jan 2003 19:48:34 -0800 Subject: [SciPy-user] ScanMail Message: To Recipient file blocking settings matched and action taken. Message-ID: <000b01c2bc48$fa36d580$075a5e40@pivia.com> ScanMail for Microsoft Exchange has blocked an attachment. Sender = big at boss.com Recipient(s) = scipy-user at scipy.net Subject = [SciPy-user] Re: Sample Scanning Time = 01/14/2003 19:48:33 Action on file blocking: The attachment Sample.pif matches the file blocking settings. ScanMail has Moved it. The attachment was moved to C:\Quarantine\Sample3e24da113.pif_. From Administrator at scipy.com Tue Jan 14 22:49:30 2003 From: Administrator at scipy.com (Administrator at scipy.com) Date: Tue, 14 Jan 2003 19:49:30 -0800 Subject: [SciPy-user] ScanMail Message: To Recipient file blocking settings matched and action taken. Message-ID: <000f01c2bc49$1be1e9e0$075a5e40@pivia.com> ScanMail for Microsoft Exchange has blocked an attachment. Sender = big at boss.com Recipient(s) = SciPy-user at scipy.net Subject = [SciPy-user] Re: Movies Scanning Time = 01/14/2003 19:49:29 Action on file blocking: The attachment Untitled1.pif matches the file blocking settings. ScanMail has Moved it. The attachment was moved to C:\Quarantine\Untitled13e24da4a4.pif_. From Administrator at scipy.com Tue Jan 14 22:52:01 2003 From: Administrator at scipy.com (Administrator at scipy.com) Date: Wed, 15 Jan 2003 04:52:01 +0100 Subject: [SciPy-user] ScanMail Message: To Recipient virus found and action taken. Message-ID: <000c01c2bc49$75e631d0$0a1f9cc1@ffi.no> ScanMail for Microsoft Exchange has detected virus-infected attachment(s). Sender = scipy-user-request at scipy.net Recipient(s) = scipy-user at scipy.net Subject = SciPy-user digest, Vol 1 #292 - 1 msg Scanning Time = 01/15/2003 04:52:01 Engine/Pattern = 6.510-1002/439 Action on virus found: The message body contains WORM_SOBIG.A virus. ScanMail has deleted the message body. Advarsen til mottaker. ScanMail har funnet et virus. From Administrator at scipy.com Tue Jan 14 22:51:20 2003 From: Administrator at scipy.com (Administrator at scipy.com) Date: Wed, 15 Jan 2003 04:51:20 +0100 Subject: [SciPy-user] ScanMail Message: To Recipient virus found and action taken. Message-ID: <000901c2bc49$5d6eeb60$0a1f9cc1@ffi.no> ScanMail for Microsoft Exchange has detected virus-infected attachment(s). Sender = scipy-user-request at scipy.net Recipient(s) = scipy-user at scipy.net Subject = SciPy-user digest, Vol 1 #291 - 1 msg Scanning Time = 01/15/2003 04:51:19 Engine/Pattern = 6.510-1002/439 Action on virus found: The message body contains WORM_SOBIG.A virus. ScanMail has deleted the message body. Advarsen til mottaker. ScanMail har funnet et virus. From Kasper.Souren at ircam.fr Thu Jan 16 11:55:10 2003 From: Kasper.Souren at ircam.fr (Kasper Souren) Date: Thu, 16 Jan 2003 17:55:10 +0100 Subject: [SciPy-user] atlas troubles during install (yes I know, I am not the first one) Message-ID: <200301161755.10166.Kasper.Souren@ircam.fr> Hi, I am trying to install SciPy, but it seems blas is causing troubles: """ atlas_info: FOUND: libraries = ['lapack', 'f77blas', 'cblas', 'atlas'] library_dirs = ['/u/formes/share/lib/i686-Linux-2.4.9-13/atlas'] include_dirs = ['/u/formes/share/include/ATLAS'] /u/formes/share/bin/i686-Linux-2.4.9-13/python2.2 /net/kempul/u.kempul/formes/souren/cvs/scipy/linalg/setup_atlas_version.py build_ext --inplace --force atlas_info: NOT AVAILABLE Traceback (most recent call last): File "/net/kempul/u.kempul/formes/souren/cvs/scipy/linalg/setup_atlas_version.py", line 25, in ? setup(**configuration()) File "/net/kempul/u.kempul/formes/souren/cvs/scipy/linalg/setup_atlas_version.py", line 15, in configuration raise AtlasNotFoundError,AtlasNotFoundError.__doc__ scipy_distutils.system_info.AtlasNotFoundError: Atlas (http://math-atlas.sourceforge.net/) libraries not found. Directories to search for the libraries can be specified in the scipy_distutils/site.cfg file (section [atlas]) or by setting the ATLAS environment variable. Failed to determine ATLAS version """ fftw_info, djbfft_info and x11_info don't cause me troubles. They come up with FOUND and it seems okay. I tried to find a solution with google, but none of the suggestions found there seems to apply. ATLAS is nicely installed, and the fact that it first finds that setup.py first states that is FOUND, and then causes a problem later seems rather strange to me. bye, Kasper From pearu at cens.ioc.ee Thu Jan 16 14:08:39 2003 From: pearu at cens.ioc.ee (Pearu Peterson) Date: Thu, 16 Jan 2003 21:08:39 +0200 (EET) Subject: [SciPy-user] atlas troubles during install (yes I know, I am not the first one) In-Reply-To: <200301161755.10166.Kasper.Souren@ircam.fr> Message-ID: On Thu, 16 Jan 2003, Kasper Souren wrote: > I am trying to install SciPy, but it seems blas is causing troubles: > > """ > atlas_info: > FOUND: > libraries = ['lapack', 'f77blas', 'cblas', 'atlas'] > library_dirs = ['/u/formes/share/lib/i686-Linux-2.4.9-13/atlas'] > include_dirs = ['/u/formes/share/include/ATLAS'] > > /u/formes/share/bin/i686-Linux-2.4.9-13/python2.2 > /net/kempul/u.kempul/formes/souren/cvs/scipy/linalg/setup_atlas_version.py > build_ext --inplace --force > atlas_info: > NOT AVAILABLE > Failed to determine ATLAS version > """ > I tried to find a solution with google, but none of the suggestions found > there seems to apply. ATLAS is nicely installed, and the fact that it first > finds that setup.py first states that is FOUND, and then causes a problem > later seems rather strange to me. How did you specify the location of atlas libraries? Currently I can think of only one situation where the above behaviour could be possible: if the location of atlas libraries was specified using relative path. If this was so in your case, then try using absolute path in ATLAS environment variable. In case you didn't use ATLAS environment variable, then try ATLAS=/u/formes/share/lib/i686-Linux-2.4.9-13/atlas python setup.py build Otherwise, could you be more specific about your environment and exactly what commands were you using when trying to build scipy? In addition to possibly relevant information as specified at the end of scipy/INSTALL.txt file, could you send also the output of python -c 'from f2py2e.diagnose import run;run()' (provided that you are using recent f2py). Pearu From lep at aber.ac.uk Fri Jan 17 05:15:50 2003 From: lep at aber.ac.uk (Leighton Pritchard) Date: Fri, 17 Jan 2003 10:15:50 +0000 Subject: [SciPy-user] Installing SciPy on Windows Message-ID: <5.1.0.14.0.20030115084542.06f83bd0@pophost.aber.ac.uk> Hi all, Having successfully installed SciPy from source on Linux (Mandrake 8.2), I'm now having difficulties installing SciPy on Windows. Can anyone please help me with the following problem: My system: Windows 2K (SP2 with pre-SP4 hotfixes) Python 2.2.2 MinGW 2.0.0 Visual Studio 6 Numeric-22.0 PIII 350, 320Mb RAM fftw win32 libraries (in C:\) F2PY2e-2.32.225-1419 Using the most recent installer SciPy-0.2.0_alpha_154.4500.win32-py2.2.exe (Jan 16 release), installation proceeds smoothly, but running scipy.test() throws an error: >>> import scipy >>> scipy.test() [...] No test suite found for scipy.stats.rw creating test suite for scipy.stats.stats ........... which brings up a dialog box: "The instruction at 0x77fcce38 referenced memory at 0x00000004. The memory could not be written" - clicking OK obviously closes the Python interpreter. The reported addresses and the stage at which the scipy.test() call fails are not always consistent. I've had the tests crash out at the ...........F... stage, too. I'm having troubles installing from source on Windows, too, but I thought I'd ask for help with the binaries first. -- Dr Leighton Pritchard AMRSC T44, Cledwyn Building Institute of Biological Sciences University of Wales, Aberystwyth, SY23 3DD Tel 01970 622353 ext. 2353 PGP public key - http://www.keyserver.net (0x47B4A485) From frank_abebe at latinmail.com Fri Jan 17 06:16:14 2003 From: frank_abebe at latinmail.com (frank Abebe) Date: Fri, 17 Jan 2003 06:16:14 -0500 Subject: [SciPy-user] FROM:FRANK Message-ID: <20030117111614.74A1B3D716E@smtp.latinmail.com> FOR YOUR URGENT ATTENTION Attn: From: Engr. Frank Abebe Federal Ministry of Works & Housing Federal Secretariat Complex Ikoyi-Lagos. Dear friend, Re: Business Proposal I am Engr. Frank Abebe, of the Federal Ministry of Works & Housing (FMW&H). Before i proceed, let me inform and make you aware that no money will be require from you, or you will not spend any money of yours into this transaction.You will only be required to receive the money. I wish to solicit your partnership and assistance in prosecuting an urgent and confidential business transaction of a multi-million dollars range which we can successfully conclude in fourteen banking days. This business will be highly beneficial to both of us and we shall be working on trust and confidentiality, It is absolutely 100% risk free and cannot fail. I came to know of you in my search for a trust worthy and reliable foreigner to assist me in prosecuting this urgent business . I shall send the details of this business as soon as you acknowledge the receipt of this letter. In appreciation of your support, you should endeavour to send me a return mail. Please, you will also ignore this letter and respect my trust in you by not exposing this transaction, even if you are not interested. So that we can move forward in this Transaction. Best Regards, Engr. FRANK ABEBE NB: I repeat again that no money of yours will be spend or invested into this transaction as every arrangement have been taken care. You will only be required to receive the money. _________________________________________________________ http://www.latinmail.com. Gratuito, latino y en espa?ol. From Kasper.Souren at ircam.fr Fri Jan 17 07:59:03 2003 From: Kasper.Souren at ircam.fr (Kasper Souren) Date: Fri, 17 Jan 2003 13:59:03 +0100 Subject: [SciPy-user] atlas troubles during install (yes I know, I am not the first one) In-Reply-To: References: Message-ID: <200301171359.03211.Kasper.Souren@ircam.fr> On Thursday 16 January 2003 20:08, Pearu Peterson wrote: > How did you specify the location of atlas libraries? Hi, I used site.cfg. I also tried the ATLAS variable, but obviously I didn't specify the right directory. > Currently I can think of only one situation where the above behaviour > could be possible: if the location of atlas libraries was specified > using relative path. If this was so in your case, then try using absolute > path in ATLAS environment variable. > In case you didn't use ATLAS environment variable, then try > > ATLAS=/u/formes/share/lib/i686-Linux-2.4.9-13/atlas python setup.py build And this works! Here was the relevant part of site.cfg: [DEFAULT] library_dirs = /u/formes/share/lib/i686-Linux-2.4.9-13/:/usr/lib include_dirs = /u/formes/share/include/:/usr/include [atlas] library_dirs = /u/formes/share/lib/i686-Linux-2.4.9-13/atlas/ include_dirs = /u/formes/share/include/ATLAS/ I had also tried several variations, but without success. After the compilation succeeded I still got the 'no fastumath error', which was fixed by running setup_scipy_base.py. (http://scipy.net/pipermail/scipy-user/2002-April/000302.html) Thank you very much! -- Kasper Souren PhD student, IRCAM Analysis/Synthesis team email: Kasper.Souren at ircam.fr | phone: +33-1-4478 4875 | Fax: +33-1-4478 1540 From nwagner at mecha.uni-stuttgart.de Fri Jan 17 08:04:32 2003 From: nwagner at mecha.uni-stuttgart.de (Nils Wagner) Date: Fri, 17 Jan 2003 14:04:32 +0100 Subject: [SciPy-user] Transcendental equations Message-ID: <3E27FF60.54A3BDFF@mecha.uni-stuttgart.de> Hi all, I would like to solve the following nonlinear equation I_m(x) J_{m-1}(x)-J_m(x) I_{m-1}(x) = 0, x > 0 with scipy, where m varies between 0 and 5. J_m Bessel function of order m I_m Modified Bessel function of order m A small example program is appreciated. Thanks in advance. Nils From lep at aber.ac.uk Fri Jan 17 10:53:43 2003 From: lep at aber.ac.uk (Leighton Pritchard) Date: Fri, 17 Jan 2003 15:53:43 +0000 Subject: [SciPy-user] Installation problems on Windows Message-ID: <5.1.0.14.0.20030117155141.0283ca28@pophost.aber.ac.uk> My earlier problem still persists with the Win32 binary release from today (Fri 17/1/03): ------------------------------------------------------------------------------- >>> scipy.test() [...] No test suite found for scipy.stats.rw creating test suite for scipy.stats.stats ........... which brings up a dialog box: "The instruction at 0x77fcce38 referenced memory at 0x00000004. The memory could not be written" - clicking OK obviously closes the Python interpreter. The reported addresses and the stage at which the scipy.test() call fails are not always consistent. I've had the tests crash out at the ...........F... stage, too. ------------------------------------------------------------------------------- any help gratefully received... -- Dr Leighton Pritchard AMRSC T44, Cledwyn Building Institute of Biological Sciences University of Wales, Aberystwyth, SY23 3DD Tel 01970 622353 ext. 2353 PGP public key - http://www.keyserver.net (0x47B4A485) From pearu at cens.ioc.ee Fri Jan 17 11:40:32 2003 From: pearu at cens.ioc.ee (Pearu Peterson) Date: Fri, 17 Jan 2003 18:40:32 +0200 (EET) Subject: [SciPy-user] Installation problems on Windows In-Reply-To: <5.1.0.14.0.20030117155141.0283ca28@pophost.aber.ac.uk> Message-ID: On Fri, 17 Jan 2003, Leighton Pritchard wrote: > My earlier problem still persists with the Win32 binary release from today > (Fri 17/1/03): > > ------------------------------------------------------------------------------- > > >>> scipy.test() > [...] > No test suite found for scipy.stats.rw creating test suite for > scipy.stats.stats > ........... > > which brings up a dialog box: "The instruction at 0x77fcce38 referenced > memory at 0x00000004. The memory could not be written" - clicking OK > obviously closes the Python interpreter. > > The reported addresses and the stage at which the scipy.test() call fails > are not always consistent. I've had the tests crash out at the > > ...........F... > > stage, too. > > ------------------------------------------------------------------------------- > > any help gratefully received... I don't have Windows machine to test this but you could try to locate which test fails by running scipy module tests one by one: >>> import scipy >>> scipy.linalg.test() >>> scipy.fftpack.test() >>> scipy.stats.test() etc. After you find which particular module (it looks like stats module) is causing the above problem, start uncommending tests cases from scipy//tests/test_.py files (see test_suite function) until you find which particular test is causing the problem. Then we can look more closely what could be the cause of these errors.. Pearu From pearu at cens.ioc.ee Fri Jan 17 11:48:32 2003 From: pearu at cens.ioc.ee (Pearu Peterson) Date: Fri, 17 Jan 2003 18:48:32 +0200 (EET) Subject: [SciPy-user] atlas troubles during install (yes I know, I am not the first one) In-Reply-To: <200301171359.03211.Kasper.Souren@ircam.fr> Message-ID: On Fri, 17 Jan 2003, Kasper Souren wrote: > On Thursday 16 January 2003 20:08, Pearu Peterson wrote: > > How did you specify the location of atlas libraries? > > I used site.cfg. I also tried the ATLAS variable, but obviously I didn't > specify the right directory. > > > Currently I can think of only one situation where the above behaviour > > could be possible: if the location of atlas libraries was specified > > using relative path. If this was so in your case, then try using absolute > > path in ATLAS environment variable. > > In case you didn't use ATLAS environment variable, then try > > > > ATLAS=/u/formes/share/lib/i686-Linux-2.4.9-13/atlas python setup.py build > > And this works! Nice. > Here was the relevant part of site.cfg: > > [DEFAULT] > library_dirs = /u/formes/share/lib/i686-Linux-2.4.9-13/:/usr/lib > include_dirs = /u/formes/share/include/:/usr/include > > [atlas] > library_dirs = /u/formes/share/lib/i686-Linux-2.4.9-13/atlas/ > include_dirs = /u/formes/share/include/ATLAS/ > > I had also tried several variations, but without success. The above site.cfg looks good and system_info.py found these atlas libraries correctly in the first case. However, when running python inside linalg library, then site.cfg was not found by system_info.py (that, I guess, was picked up from installed scipy_distutils instead of cvs/scipy/scipy_distutils) and therefore atlas libraries were not found in the second case. I'll look how to fix this later.. until then the site.cfg feature regarding specifying atlas libraries in site.cfg file is broken. Pearu From nwagner at mecha.uni-stuttgart.de Fri Jan 17 11:51:12 2003 From: nwagner at mecha.uni-stuttgart.de (Nils Wagner) Date: Fri, 17 Jan 2003 17:51:12 +0100 Subject: [SciPy-user] Problems with python setup.py install (latest cvs) Message-ID: <3E283480.8E73C1E5@mecha.uni-stuttgart.de> Hi all, I have used latest cvs to install scipy. However python setup.py install --force failed. This is the output raceback (most recent call last): File "setup.py", line 127, in ? install_package() File "setup.py", line 117, in install_package url = "http://www.scipy.org", File "scipy_distutils/core.py", line 42, in setup return old_setup(**new_attr) File "/usr/local/lib/python2.1/distutils/core.py", line 138, in setup dist.run_commands() File "/usr/local/lib/python2.1/distutils/dist.py", line 899, in run_commands self.run_command(cmd) File "/usr/local/lib/python2.1/distutils/dist.py", line 919, in run_command cmd_obj.run() File "/usr/local/lib/python2.1/distutils/command/install.py", line 484, in run self.run_command(cmd_name) File "/usr/local/lib/python2.1/distutils/cmd.py", line 328, in run_command self.distribution.run_command(command) File "/usr/local/lib/python2.1/distutils/dist.py", line 919, in run_command cmd_obj.run() File "/usr/local/lib/python2.1/distutils/command/install_lib.py", line 93, in run self.byte_compile(outfiles) File "/usr/local/lib/python2.1/distutils/command/install_lib.py", line 130, in byte_compile verbose=self.verbose, dry_run=self.dry_run) File "/usr/local/lib/python2.1/distutils/util.py", line 439, in byte_compile compile(file, cfile, dfile) File "/usr/local/lib/python2.1/py_compile.py", line 62, in compile codeobject = __builtin__.compile(codestring, dfile or file, 'exec') OverflowError: ('integer literal too large', ('/usr/local/lib/python2.1/site-packages/scipy/special/tests/test_special.py', 357, None, 'assert_almost_equal(eu2[20],370371188237525,0)\n')) Any idea ? Nils From oliphant.travis at ieee.org Fri Jan 17 16:33:00 2003 From: oliphant.travis at ieee.org (Travis Oliphant) Date: 17 Jan 2003 14:33:00 -0700 Subject: [SciPy-user] Transcendental equations In-Reply-To: <3E27FF60.54A3BDFF@mecha.uni-stuttgart.de> References: <3E27FF60.54A3BDFF@mecha.uni-stuttgart.de> Message-ID: <1042839182.25192.3.camel@travis.local.net> On Fri, 2003-01-17 at 06:04, Nils Wagner wrote: > Hi all, > > I would like to solve the following nonlinear equation > > I_m(x) J_{m-1}(x)-J_m(x) I_{m-1}(x) = 0, x > 0 > > with scipy, where m varies between 0 and 5. You can use any of the zero finding routines, except you have to know something about which zero you want to find, as this particular function has many zeros. For example, to find the zero-crossing nearest to x=0, you can use the following code: import scipy iv = scipy.special.iv jv = scipy.special.jv solver = scipy.optimize.brentq def zerofunc(x,m): return iv(m,x)*jv(m-1,x)-jv(m,x)*iv(m-1,x) for m in range(0,6): endint = 5 + m root = solver(zerofunc,1e-4,endint,args=(m,)) print "m=%d, root=%10.8f, function value=%10.8f" % \ (m, root, zerofunc(root,m)) The output of this code is m=0, root=3.19622062, function value=-0.00000000 m=1, root=4.61089988, function value=0.00000000 m=2, root=5.90567824, function value=0.00000000 m=3, root=7.14353102, function value=0.00000000 m=4, root=8.34660594, function value=0.00000000 m=5, root=9.52570136, function value=0.00000000 -Travis O. From baecker at physik.tu-dresden.de Sun Jan 19 04:58:50 2003 From: baecker at physik.tu-dresden.de (baecker at physik.tu-dresden.de) Date: Sun, 19 Jan 2003 10:58:50 +0100 (CET) Subject: [SciPy-user] dynamic and interactive 2D ? Message-ID: Hi, I am wondering if there is a package which allows for dynamic representation of data + mouse events ? For example I would like to plot iterates of an initial point for a 2 dimensional map until the user selects via mouse a new initial point. This may lead to several thousands of points to be plotted dynamically in a graphics window. (Storing them is not necessary.) Would you recommend direct coding with a tkinter canvas? Any pointers are welcome!! ((Background Info: we are presently considering to use Python+Numeric+scipy together with some good graphics add-on for a course on computational physics next summer term. One crucial aspect for the choice of programming language/system is dynamic display of data + interactive aspects via mouse ... Currently PVWave comes closest to this, but I really don't like the language (+cost) and I'd rather very much prefer python ..)) Many thanks, Arnd From jsteve17 at tampabay.rr.com Sun Jan 19 09:55:46 2003 From: jsteve17 at tampabay.rr.com (Jeffrey Stephens) Date: Sun, 19 Jan 2003 09:55:46 -0500 Subject: [SciPy-user] Error Compiling Scipy Message-ID: <200301190955.46058.jsteve17@tampabay.rr.com> I am running RH 8.0, KDE Desktop, and using Python 2.2. I began compiling Scipy, and everything went fine for a half hour or so. Then the compile aborted with the following message. g77 -shared build/temp.linux-i686-2.2/fftpackmodule.o -Lbuild/temp.linux-i686-2.2 -Lbuild/temp.linux-i686-2.2 -lfftpack -ldfftpack -lg2c -o build/lib.linux-i686-2.2/scipy/fftpack/fftpack.so restoring linker_so 'gcc -shared' building 'scipy.xplt.gistC' extension error: file '/home/jeffs/update/SciPy-0.2.0_alpha_144.4350/xplt/gistCmodule.c' does not exist A search reveals there is no gistCmodule.c anywhere in in SciPy. Is this a bug? Need help? Regards, JeffS From prabhu at aero.iitm.ernet.in Sun Jan 19 11:39:27 2003 From: prabhu at aero.iitm.ernet.in (Prabhu Ramachandran) Date: Sun, 19 Jan 2003 22:09:27 +0530 Subject: [SciPy-user] dynamic and interactive 2D ? In-Reply-To: References: Message-ID: <15914.54463.780480.664341@monster.linux.in> >>>>> "AB" == baecker writes: AB> Hi, I am wondering if there is a package which allows for AB> dynamic representation of data + mouse events ? For example I AB> would like to plot iterates of an initial point for a 2 AB> dimensional map until the user selects via mouse a new initial AB> point. This may lead to several thousands of points to be AB> plotted dynamically in a graphics window. (Storing them is not AB> necessary.) Would you recommend direct coding with a tkinter AB> canvas? Depending on the GUI toolkit you choose there should be many ways of doing this. Pick a toolkit and then pick a plotting widget to do the job. If you choose Qt, there is a Qt based graph widget called PyQwt (http://gerard.vermeulen.free.fr). If you choose wxPython you could use scipy's plt module or better still chaco. If you want to use Tkinter you can use Pmw.Blt.Graph or use Tkinters canvas directly. If you are going to use pyGTK I think there is a plot widget for it too. So my suggestion is that you pick a GUI toolkit first. :) cheers, prabhu From pearu at cens.ioc.ee Sun Jan 19 11:56:45 2003 From: pearu at cens.ioc.ee (Pearu Peterson) Date: Sun, 19 Jan 2003 18:56:45 +0200 (EET) Subject: [SciPy-user] Error Compiling Scipy In-Reply-To: <200301190955.46058.jsteve17@tampabay.rr.com> Message-ID: On Sun, 19 Jan 2003, Jeffrey Stephens wrote: > I am running RH 8.0, KDE Desktop, and using Python 2.2. I began compiling > Scipy, and everything went fine for a half hour or so. Then the compile > aborted with the following message. > > g77 -shared build/temp.linux-i686-2.2/fftpackmodule.o > -Lbuild/temp.linux-i686-2.2 -Lbuild/temp.linux-i686-2.2 -lfftpack -ldfftpack > -lg2c -o build/lib.linux-i686-2.2/scipy/fftpack/fftpack.so > restoring linker_so 'gcc -shared' > building 'scipy.xplt.gistC' extension > error: file '/home/jeffs/update/SciPy-0.2.0_alpha_144.4350/xplt/gistCmodule.c' > does not exist > > A search reveals there is no gistCmodule.c anywhere in in SciPy. Is this a > bug? Need help? The reason why given Scipy tarball lacks gistCmodule.c, in fact the whole module is missing, is that it was generated under windows system that skips xplt module. Currently I recommend using Scipy from its CVS repository. Or you can disable xplt module from scipy setup.py file in order to finish scipy build. Pearu From jsteve17 at tampabay.rr.com Sun Jan 19 17:00:24 2003 From: jsteve17 at tampabay.rr.com (Jeffrey Stephens) Date: Sun, 19 Jan 2003 17:00:24 -0500 Subject: [SciPy-user] Error Compiling Scipy In-Reply-To: References: Message-ID: <200301191700.24583.jsteve17@tampabay.rr.com> help? > > The reason why given Scipy tarball lacks gistCmodule.c, in fact the > whole module is missing, is that it was generated under windows system > that skips xplt module. > Currently I recommend using Scipy from its CVS repository. > Or you can disable xplt module from scipy setup.py file in order to finish > scipy build. > > Pearu > Thanks. I commented out the code, and the compile finished. I just tried running the tests, and got some errors followed by a segmentation fault. I think I may have hosed the installation, because I first installed the from binary, but got an immediate segmentation fault as soon as executed: >>> from scipy import * So then I downloaded the source tarball and installed that over the binary. What would you recommend? Should I delete everything, and start over? If so, should I download from CVS instead of using the source tarball? Also, I didn't pay any attention to where the installed files went. Can they be conveniently deleted by removing a directory or directories? Thanks. Regards, Jeff S From pearu at cens.ioc.ee Sun Jan 19 17:18:51 2003 From: pearu at cens.ioc.ee (Pearu Peterson) Date: Mon, 20 Jan 2003 00:18:51 +0200 (EET) Subject: [SciPy-user] Error Compiling Scipy In-Reply-To: <200301191700.24583.jsteve17@tampabay.rr.com> Message-ID: On Sun, 19 Jan 2003, Jeffrey Stephens wrote: > Thanks. I commented out the code, and the compile finished. I just tried > running the tests, and got some errors followed by a segmentation fault. > I think I may have hosed the installation, because I first installed the from > binary, but got an immediate segmentation fault as soon as executed: > > >>> from scipy import * > > So then I downloaded the source tarball and installed that over the binary. > What would you recommend? Should I delete everything, and start over? > If so, should I download from CVS instead of using the source tarball? Also, > I didn't pay any attention to where the installed files went. Can they be > conveniently deleted by removing a directory or directories? Thanks. Yes, starting over from deleting everything and downloading scipy from CVS might be a good idea. Packages related to Scipy are: scipy_base, gui_thread, scipy_distutils, scipy, and scipy_test. They all are installed to /lib/python/site-packages directory where you can remove. To build scipy, run python setup_scipy_core.py install python setup_scipy.py install in scipy CVS directory. Pearu From lep at aber.ac.uk Mon Jan 20 05:12:33 2003 From: lep at aber.ac.uk (Leighton Pritchard) Date: Mon, 20 Jan 2003 10:12:33 +0000 Subject: [SciPy-user] Re: SciPy-user digest, Vol 1 #295 - 11 msgs In-Reply-To: <20030117180003.14202.73335.Mailman@shaft> Message-ID: <5.1.0.14.0.20030120095735.02837008@pophost.aber.ac.uk> At 12:00 17/01/2003, you wrote: >On Fri, 17 Jan 2003, Leighton Pritchard wrote: > > > My earlier problem still persists with the Win32 binary release from today > > (Fri 17/1/03): [...] > > any help gratefully received... > >I don't have Windows machine to test this but you could >try to locate which test fails by running scipy module tests >one by one: > >>> import scipy > >>> scipy.linalg.test() > >>> scipy.fftpack.test() > >>> scipy.stats.test() >etc. >After you find which particular module (it looks like stats module) is >causing the above problem, start uncommending tests cases from > scipy//tests/test_.py >files (see test_suite function) until you find which particular test is >causing the problem. Then we can look more closely what could be the cause >of these errors.. Many thanks, Pearu. I have done this and the problem seems to be in fftpack/test_basic.py. The tests that fail are: test_fft test_ifft test_rfft test_irfft all with the 'check_' parameter. Tests for test_fftn test_ifftn with 'check_' run fine (the tests with the 'bench_' parameter don't seem to run any tests) Commenting out the method check_djbfft() in the test_fft, test_ifft, test_rfft and test_irfft classes allows the tests to run. The offending line in the check_djbfft() method appears to be: y2 = fft2(x) Commenting this line out (and subsequent references to y2) allows the tests to run. I hope this helps - if you need more details about the problem, please let me know. -- Dr Leighton Pritchard AMRSC T44, Cledwyn Building Institute of Biological Sciences University of Wales, Aberystwyth, SY23 3DD Tel 01970 622353 ext. 2353 PGP public key - http://www.keyserver.net (0x47B4A485) From pearu at cens.ioc.ee Mon Jan 20 05:28:58 2003 From: pearu at cens.ioc.ee (Pearu Peterson) Date: Mon, 20 Jan 2003 12:28:58 +0200 (EET) Subject: [SciPy-user] Re: SciPy-user digest, Vol 1 #295 - 11 msgs In-Reply-To: <5.1.0.14.0.20030120095735.02837008@pophost.aber.ac.uk> Message-ID: On Mon, 20 Jan 2003, Leighton Pritchard wrote: > (the tests with the 'bench_' parameter don't seem to run any tests) 'bench_' prefixed test methods are run when the testing level is higher than 5, default is 1. You can specify higher testing level as follows: >>> scipy.test(level=10) for instance. > Commenting out the method check_djbfft() in the test_fft, test_ifft, > test_rfft and test_irfft classes allows the tests to run. > > The offending line in the check_djbfft() method appears to be: > > y2 = fft2(x) Hmm, this fft2 refers to a Numeric FFT function. What version of Numeric are you using? Is FFT from Numeric working properly in your case? Try for example from FFT import fft a=fft(range(2**4)) > Commenting this line out (and subsequent references to y2) allows the tests > to run. > > I hope this helps - if you need more details about the problem, please let > me know. Thanks for feedback! Pearu From nwagner at mecha.uni-stuttgart.de Mon Jan 20 07:39:49 2003 From: nwagner at mecha.uni-stuttgart.de (Nils Wagner) Date: Mon, 20 Jan 2003 13:39:49 +0100 Subject: [SciPy-user] Transcendental equations References: <3E27FF60.54A3BDFF@mecha.uni-stuttgart.de> <1042839182.25192.3.camel@travis.local.net> Message-ID: <3E2BEE15.9455F668@mecha.uni-stuttgart.de> Travis Oliphant schrieb: > > On Fri, 2003-01-17 at 06:04, Nils Wagner wrote: > > Hi all, > > > > I would like to solve the following nonlinear equation > > > > I_m(x) J_{m-1}(x)-J_m(x) I_{m-1}(x) = 0, x > 0 > > > > with scipy, where m varies between 0 and 5. > > You can use any of the zero finding routines, except you have to know > something about which zero you want to find, as this particular function > has many zeros. > > For example, to find the zero-crossing nearest to x=0, you can use the > following code: > > import scipy > > iv = scipy.special.iv > jv = scipy.special.jv > solver = scipy.optimize.brentq > > def zerofunc(x,m): > return iv(m,x)*jv(m-1,x)-jv(m,x)*iv(m-1,x) > > for m in range(0,6): > endint = 5 + m > root = solver(zerofunc,1e-4,endint,args=(m,)) > print "m=%d, root=%10.8f, function value=%10.8f" % \ > (m, root, zerofunc(root,m)) > > The output of this code is > > m=0, root=3.19622062, function value=-0.00000000 > m=1, root=4.61089988, function value=0.00000000 > m=2, root=5.90567824, function value=0.00000000 > m=3, root=7.14353102, function value=0.00000000 > m=4, root=8.34660594, function value=0.00000000 > m=5, root=9.52570136, function value=0.00000000 > > -Travis O. > > _______________________________________________ > SciPy-user mailing list > SciPy-user at scipy.net > http://www.scipy.net/mailman/listinfo/scipy-user Travis, Thank you for your prompt reply. Actually, I am interested in finding all roots for a given interval, let's say [0,a] of the transcendental equation for fixed m. How can I solve this problem ? How can I guarantee that I don't miss out any zero in that interval ? Nils From lep at aber.ac.uk Mon Jan 20 09:40:52 2003 From: lep at aber.ac.uk (Leighton Pritchard) Date: Mon, 20 Jan 2003 14:40:52 +0000 Subject: [SciPy-user] Windows binary installation problems Message-ID: <5.1.0.14.0.20030120141333.027e23a0@pophost.aber.ac.uk> Pearu wrote: >On Mon, 20 Jan 2003, Leighton Pritchard wrote: > > (the tests with the 'bench_' parameter don't seem to run any tests) >'bench_' prefixed test methods are run when the testing level is higher >than 5, default is 1. You can specify higher testing level as follows: > >>> scipy.test(level=10) for instance. I was just hitting 'F5' in IDLEfork for the individual tests, to save me having to restart the command-line over and over in Windows. I should have been paying more attention to the code ;) > > Commenting out the method check_djbfft() in the test_fft, test_ifft, > > test_rfft and test_irfft classes allows the tests to run. > > > > The offending line in the check_djbfft() method appears to be: > > > > y2 = fft2(x) >Hmm, this fft2 refers to a Numeric FFT function. What version of Numeric >are you using? Numeric 22.0, installed from the binaries on Sourceforge. >Is FFT from Numeric working properly in your case? Try for example from FFT >import fft >a=fft(range(2**4)) I got odd(-ish) behaviour with this. This example for fft seemed to work fine in command-line Python, but it caused the crashing behaviour in IDLEfork, while 'ordinary' IDLE occasionally let me type characters on the next line before crashing. Very confusing... I cleared out Numeric and old associated packages, then reinstalled it... now everything works in the cmd-line/IDLE interpreter window(s) and in IDLEfork. A reinstall of SciPy passes all the SciPy tests in the two interpreters and IDLEfork. Fantastic! I think I left the residue of an obsolete FFT package from an old installation of Numeric, and this was causing the problem. Once I'd cleaned it out and reinstalled, everything was fine with SciPy. Many thanks for your help and patience, -- Dr Leighton Pritchard AMRSC T44, Cledwyn Building Institute of Biological Sciences University of Wales, Aberystwyth, SY23 3DD Tel 01970 622353 ext. 2353 PGP public key - http://www.keyserver.net (0x47B4A485) From oliphant at ee.byu.edu Mon Jan 20 16:25:18 2003 From: oliphant at ee.byu.edu (Travis Oliphant) Date: Mon, 20 Jan 2003 14:25:18 -0700 (MST) Subject: [SciPy-user] dynamic and interactive 2D ? In-Reply-To: Message-ID: > Hi, > > I am wondering if there is a package which allows > for dynamic representation of data + mouse events ? > For example I would like to plot > iterates of an initial point for a 2 dimensional map until > the user selects via mouse a new initial point. > This may lead to several thousands of points to be plotted > dynamically in a graphics window. (Storing them is not necessary.) > Would you recommend direct coding with a tkinter canvas? > > Any pointers are welcome!! > Right now, xplt on Unix will get you part of the way there. If you are not on Unix, then Chaco is getting pretty far along. -Travis O. -- Travis Oliphant Assistant Professor 459 CB Electrical and Computer Engineering Brigham Young University Provo, UT 84602 Tel: (801) 422-3108 oliphant.travis at ieee.org From oliphant at ee.byu.edu Mon Jan 20 16:29:49 2003 From: oliphant at ee.byu.edu (Travis Oliphant) Date: Mon, 20 Jan 2003 14:29:49 -0700 (MST) Subject: [SciPy-user] Error Compiling Scipy In-Reply-To: <200301190955.46058.jsteve17@tampabay.rr.com> Message-ID: > I am running RH 8.0, KDE Desktop, and using Python 2.2. I began compiling > Scipy, and everything went fine for a half hour or so. Then the compile > aborted with the following message. > > g77 -shared build/temp.linux-i686-2.2/fftpackmodule.o > -Lbuild/temp.linux-i686-2.2 -Lbuild/temp.linux-i686-2.2 -lfftpack -ldfftpack > -lg2c -o build/lib.linux-i686-2.2/scipy/fftpack/fftpack.so > restoring linker_so 'gcc -shared' > building 'scipy.xplt.gistC' extension > error: file '/home/jeffs/update/SciPy-0.2.0_alpha_144.4350/xplt/gistCmodule.c' > does not exist > > A search reveals there is no gistCmodule.c anywhere in in SciPy. Is this a > bug? Need help? > This is a bug in how SciPy is built (if the source package is created on a Windows computer it doesn't include this Unix-only package. You can check out the package from scipy CVS). This is something that needs to be fixed. -Travis -- Travis Oliphant Assistant Professor 459 CB Electrical and Computer Engineering Brigham Young University Provo, UT 84602 Tel: (801) 422-3108 oliphant.travis at ieee.org From oliphant at ee.byu.edu Mon Jan 20 16:35:50 2003 From: oliphant at ee.byu.edu (Travis Oliphant) Date: Mon, 20 Jan 2003 14:35:50 -0700 (MST) Subject: [SciPy-user] Transcendental equations In-Reply-To: <3E2BEE15.9455F668@mecha.uni-stuttgart.de> Message-ID: > Travis, > > Thank you for your prompt reply. Actually, I am interested > in finding all roots for a given interval, let's say [0,a] of the > transcendental > equation for fixed m. How can I solve this problem ? > How can I guarantee that I don't miss out any zero in that interval ? I don't know, exactly. Without knowing something about the function, I don't know how you can guarantee you don't miss a zero. You could try to fit the function to a polynomial and use the zeros of that polynomial as a starting point -- then you have to pick the order. You could also just try some kind of brute force method, where you break up the interval into smaller and smaller pieces until you have bracketed all the zeros. If anybody knows of a better idea, please let me know. I might try to do something along the lines of the bracketing idea. -Travis -- Travis Oliphant Assistant Professor 459 CB Electrical and Computer Engineering Brigham Young University Provo, UT 84602 Tel: (801) 422-3108 oliphant.travis at ieee.org From nwagner at mecha.uni-stuttgart.de Tue Jan 21 07:08:39 2003 From: nwagner at mecha.uni-stuttgart.de (Nils Wagner) Date: Tue, 21 Jan 2003 13:08:39 +0100 Subject: [SciPy-user] Transcendental equations References: Message-ID: <3E2D3847.9EC26027@mecha.uni-stuttgart.de> Travis Oliphant schrieb: > > > Travis, > > > > Thank you for your prompt reply. Actually, I am interested > > in finding all roots for a given interval, let's say [0,a] of the > > transcendental > > equation for fixed m. How can I solve this problem ? > > How can I guarantee that I don't miss out any zero in that interval ? > > I don't know, exactly. Without knowing something about the function, I > don't know how you can guarantee you don't miss a zero. > > You could try to fit the function to a polynomial and use the zeros of > that polynomial as a starting point -- then you have to pick the order. > > You could also just try some kind of brute force method, where you break > up the interval into smaller and smaller pieces until you have bracketed > all the zeros. > > If anybody knows of a better idea, please let me know. I might try to do > something along the lines of the bracketing idea. > > -Travis > In this special case, it can be shown that the solutions of the transcendental equation I_m(x_n) J_{m-1}(x_n)-J_m(x_n) I_{m-1}(x_n) = 0, x > 0, m = 0,1,2,... n = 1,2,... tend to x_{mn} = (0.5*m+n) \pi for large n. This might be useful in finding good starting values x_0. Nils > -- > Travis Oliphant > Assistant Professor > 459 CB > Electrical and Computer Engineering > Brigham Young University > Provo, UT 84602 > Tel: (801) 422-3108 > oliphant.travis at ieee.org > > _______________________________________________ > SciPy-user mailing list > SciPy-user at scipy.net > http://www.scipy.net/mailman/listinfo/scipy-user From lac at strakt.com Tue Jan 21 13:46:35 2003 From: lac at strakt.com (Laura Creighton) Date: Tue, 21 Jan 2003 19:46:35 +0100 Subject: [SciPy-user] suggestion for improving http://www.scipy.org/ Message-ID: <200301211846.h0LIkZkU013042@ratthing-b246.strakt.com> Recently I wanted to find out 'Can scipy generate a random variable from a power law distribution'. So I went to the site, and searched for 'power law random'. Yow. 445 hits, none of which appear to have anything to do with generating random numbers, most of which are the empty home pages of people like me who have just joined the community. Searching for 'power and law' did not help. So -- 1. is the search engine broken, or do you all know something I don't? 2. Any chance of implementing a specialised search for 'Can Scipy do this?' Thanks very much for your time, and for scipy, Laura Creighton From travis at enthought.com Tue Jan 21 15:49:50 2003 From: travis at enthought.com (Travis N. Vaught) Date: Tue, 21 Jan 2003 14:49:50 -0600 Subject: [SciPy-user] suggestion for improving http://www.scipy.org/ In-Reply-To: <200301211846.h0LIkZkU013042@ratthing-b246.strakt.com> Message-ID: <00c401c2c18e$a451b450$0300a8c0@tvlaptop> The web site search engine is currently broken--not sure when it happened, but it's been down a few months. I wasn't sure if anyone was really using it--I'm sure no one is if it's broken. Searching the mailing lists works--but probably won't give you what you want. I'll look at fixing the catalog or whatever is causing the zope search to return every doc. Thanks Travis > -----Original Message----- > From: scipy-user-admin at scipy.net [mailto:scipy-user-admin at scipy.net] On > Behalf Of Laura Creighton > Sent: Tuesday, January 21, 2003 12:47 PM > To: scipy-user at scipy.net > Subject: [SciPy-user] suggestion for improving http://www.scipy.org/ > > Recently I wanted to find out 'Can scipy generate a random variable > from a power law distribution'. So I went to the site, and > searched for 'power law random'. Yow. 445 hits, none of which > appear to have anything to do with generating random numbers, > most of which are the empty home pages of people like me who have > just joined the community. Searching for 'power and law' did not > help. > > So -- 1. is the search engine broken, or do you all know something I > don't? 2. Any chance of implementing a specialised search for > 'Can Scipy do this?' > > Thanks very much for your time, > and for scipy, > Laura Creighton > _______________________________________________ > SciPy-user mailing list > SciPy-user at scipy.net > http://www.scipy.net/mailman/listinfo/scipy-user From otttr440 at student.liu.se Tue Jan 21 17:03:27 2003 From: otttr440 at student.liu.se (Otto Tronarp) Date: 21 Jan 2003 23:03:27 +0100 Subject: [SciPy-user] suggestion for improving http://www.scipy.org/ In-Reply-To: <200301211846.h0LIkZkU013042@ratthing-b246.strakt.com> References: <200301211846.h0LIkZkU013042@ratthing-b246.strakt.com> Message-ID: <1043186606.11632.5.camel@mathcore2> I don't know if you did found out if there is a power law dist. However, a quick grep in the source tree gives this function in scipy/stats/distributions.py ## Power-function distribution ## Special case of beta dist. with d =1.0 class powerlaw_gen(rv_continuous): def _pdf(self, x, a): return a*x**(a-1.0) def _cdf(self, x, a): return x**(a*1.0) def _ppf(self, q, a): return pow(q, 1.0/a) def _stats(self, a): return a/(a+1.0), a*(a+2.0)/(a+1.0)**2, \ 2*(1.0-a)*sqrt((a+2.0)/(a*(a+3.0))), \ 6*polyval([1,-1,-6,2],a)/(a*(a+3.0)*(a+4)) powerlaw = powerlaw_gen(a=0.0, b=1.0) I don't know if that's what you want, but at least it sounds like the right thing. Regards, Otto On Tue, 2003-01-21 at 19:46, Laura Creighton wrote: > Recently I wanted to find out 'Can scipy generate a random variable > from a power law distribution'. So I went to the site, and > searched for 'power law random'. Yow. 445 hits, none of which > appear to have anything to do with generating random numbers, > most of which are the empty home pages of people like me who have > just joined the community. Searching for 'power and law' did not > help. > > So -- 1. is the search engine broken, or do you all know something I > don't? 2. Any chance of implementing a specialised search for > 'Can Scipy do this?' > > Thanks very much for your time, > and for scipy, > Laura Creighton > _______________________________________________ > SciPy-user mailing list > SciPy-user at scipy.net > http://www.scipy.net/mailman/listinfo/scipy-user -- Otto Tronarp From oliphant at ee.byu.edu Tue Jan 21 17:04:12 2003 From: oliphant at ee.byu.edu (Travis Oliphant) Date: Tue, 21 Jan 2003 15:04:12 -0700 (MST) Subject: [SciPy-user] suggestion for improving http://www.scipy.org/ In-Reply-To: <200301211846.h0LIkZkU013042@ratthing-b246.strakt.com> Message-ID: > Recently I wanted to find out 'Can scipy generate a random variable > from a power law distribution'. So I went to the site, and > searched for 'power law random'. Yow. 445 hits, none of which > appear to have anything to do with generating random numbers, > most of which are the empty home pages of people like me who have > just joined the community. Searching for 'power and law' did not > help. > > So -- 1. is the search engine broken, or do you all know something I > don't? 2. Any chance of implementing a specialised search for > 'Can Scipy do this?' good idea. A start in this direction is in SciPy. If you type info('powerlaw') you get a reference to the stats module. info(stats.powerlaw) shows that this is a supported distribution. Unfortunately, the docstring is not very informative due to the way I've reconstructed the stats module to inherit from base classes (anybody who could help me figure out how to make the doc strings more informative would be appreciated). For now you have to look at the continuous_distributions.lyx file for information about all the supported distributions. That file is quite complete, and we should make an html page or a pdf file out of it. Thanks for your input. -Travis O. -- Travis Oliphant Assistant Professor 459 CB Electrical and Computer Engineering Brigham Young University Provo, UT 84602 Tel: (801) 422-3108 oliphant.travis at ieee.org From yann.ledu at noos.fr Tue Jan 21 19:33:14 2003 From: yann.ledu at noos.fr (Yann Le Du) Date: Wed, 22 Jan 2003 01:33:14 +0100 (CET) Subject: [SciPy-user] Using the mouse in gplt ? In-Reply-To: <000001c2af7b$bf4c44c0$8901a8c0@ERICDESKTOP> Message-ID: On Sun, 29 Dec 2002, eric jones wrote: > You can send raw gnuplot commands to a plot object like so: > > >>> from scipy import gplt > >>> gplt.plot((1,2,3)) > >>> q=gplt.current() > >>> q > > >>> q._send('set mouse;') > > eric Thank you : it works for me if I first send 'set mouse' BEFORE I use gplt.plot(). I found this thanks to Arnd's email, saying that 'set mouse' should be sent as the first command. Many thanks for that great tip ! Yann From yann.ledu at noos.fr Tue Jan 21 19:35:53 2003 From: yann.ledu at noos.fr (Yann Le Du) Date: Wed, 22 Jan 2003 01:35:53 +0100 (CET) Subject: [SciPy-user] Using the mouse in gplt ? In-Reply-To: Message-ID: On Sun, 29 Dec 2002 baecker at physik.tu-dresden.de wrote: > Hi, > > to answer your question partially: you have to issue > the command "set mouse" to gnuplot (as first command you send to gnuplot). > However, I don't know how to do this within scipy.plt. I used some of your infos with Eric Jones infos and succeeded ! The thing is to send 'set mouse' first, as you noted. Thanks for the help ! And I do use IPython, only I use it in the -p scipy mode and I wished to use scipy's gplt command, so I'm happy now ! Yann > > Alternatively, you might consider IPython, > http://www-hep.colorado.edu/~fperez/ipython/ > which I would highly recommend for interactive work. > Just start it with the numeric profile, > ipython -profile numeric > and then at the prompt do (for example) > g("plot sin(x)") > From edouard.oudet at wanadoo.fr Wed Jan 22 10:02:56 2003 From: edouard.oudet at wanadoo.fr (Utilisateur principal) Date: Wed, 22 Jan 2003 14:02:56 -0100 Subject: [SciPy-user] Install CVS on Mandrake i686 Message-ID: <000a01c2c227$59f47020$0fb60f50@oemcomputer> I tried to install the CVS version of Scipy on Mabdrake 8.2, arch=i686. After the following steps (suggested by Fernando Perez http://www.scipy.org/Members/fperez/Perez/PerezCVSBuild.htm ): - installing Python+Numerci, - installing fftw libraries+ extra copy - installing lapack and blas rpms : blas-3.0-18.i386.rpm and lapack-3.0-9.i386.rpm, - installing atlas-3.4.1-4mdk.noarch.rpm + succesfull tests, - creating a complete atlas-lapack library as suggested on atlas web page, - installing f2py. I tried to install scipy CVS with the command >python setup.py build I first get some problems with the package xplt. I decided to comment the line on this subject in the file setup.py. Now it is possible to build and install scipy. Furthermore, some packages work (like optimization). Unfortunately, the command >scipy.test() gave the lines No test suite found for scipy.io creating test suite for: scipy.linalg.basic Erreur de segmentation As an example, it is of course to find eigenvalues of a matrix (for instance with linalg.eig(rand(3,3))) : it gives File ... File ... File ... ImportError: /usr/lib/python2.2/site-packages/scipy/linalg/flapack.so undefined symbol : min Is it a problem with my rpm's of lapack ? Does anyone have an idea ?? Thanks a lot for help !!!!!!! -------------- next part -------------- An HTML attachment was scrubbed... URL: From pearu at cens.ioc.ee Wed Jan 22 08:38:04 2003 From: pearu at cens.ioc.ee (Pearu Peterson) Date: Wed, 22 Jan 2003 15:38:04 +0200 (EET) Subject: [SciPy-user] Install CVS on Mandrake i686 In-Reply-To: <000a01c2c227$59f47020$0fb60f50@oemcomputer> Message-ID: On Wed, 22 Jan 2003, Utilisateur principal wrote: > I tried to install scipy CVS with the command > > >python setup.py build > > I first get some problems with the package xplt. I decided to comment the line on this subject in the file setup.py. > Now it is possible to build and install scipy. Furthermore, some packages work (like optimization). Unfortunately, the command > > >scipy.test() > > gave the lines > > No test suite found for scipy.io > creating test suite for: scipy.linalg.basic > Erreur de segmentation > > As an example, it is of course to find eigenvalues of a matrix (for instance with linalg.eig(rand(3,3))) : it gives > > File ... > File ... > File ... > ImportError: /usr/lib/python2.2/site-packages/scipy/linalg/flapack.so undefined symbol : min > > Is it a problem with my rpm's of lapack ? > Does anyone have an idea ?? Thanks a lot for help !!!!!!! What f2py version are you using? Make sure it is the latest (2.32.225-1419) or from CVS. If that doesn't help, follow troubleshooting instructions from scipy/INSTALL.txt when reporting problems. Pearu From alobo at ija.csic.es Wed Jan 22 11:18:23 2003 From: alobo at ija.csic.es (Agustin Lobo) Date: Wed, 22 Jan 2003 17:18:23 +0100 (MET) Subject: [SciPy-user] Problem with plt.plot Message-ID: Hi! I'm following the tutorial "Minimal Python for Scientific Computing" and get to this problem when I try to use plt: >>> from scipy import * >>> import gui_thread >> Exception in thread Thread-1: Traceback (most recent call last): File "/var/tmp/python-root//usr/lib/python2.1/threading.py", line 378, in __bootstrap self.run() File "/var/tmp/python-root//usr/lib/python2.1/threading.py", line 366, in run apply(self.__target, self.__args, self.__kwargs) File "/tmp/SciPyTest/linux2/lib/python2.1/site-packages/gui_thread/main.py", line 38, in gui_thread File "/tmp/SciPyTest/linux2/lib/python2.1/site-packages/gui_thread/gui_thread_guts.py", line 44, in ? ImportError: No module named wxPython.wx should download and install wxPython just to use plt.plot ? Thanks Agus Dr. Agustin Lobo Instituto de Ciencias de la Tierra (CSIC) Lluis Sole Sabaris s/n 08028 Barcelona SPAIN tel 34 93409 5410 fax 34 93411 0012 alobo at ija.csic.es From oudet at irmasrv2.u-strasbg.fr Wed Jan 22 12:00:50 2003 From: oudet at irmasrv2.u-strasbg.fr (Edouard Oudet) Date: Wed, 22 Jan 2003 18:00:50 +0100 (CET) Subject: [SciPy-user] Precisions : Install CVS on Mandrake i686 In-Reply-To: Message-ID: Here are the precisions that I should give in my first email (sorry)...: ----------------------------------------------------------------------------- 1) Platform information: posix linux-i386 Linux localhost.localdomain 2.4.18-6mdk #1 Fri Mar 15 02:59:08 CET 2002 i686 unknown ----------------------------------------------------------------------------- 2)Information about C,C++,Fortran compilers/linkers Reading specs from /usr/lib/gcc-lib/i586-mandrake-linux-gnu/2.96/specs gcc version 2.96 20000731 (Mandrake Linux 8.2 2.96-0.76mdk) g77 --verbose gives g77 version 2.96 20000731 (Mandrake Linux 8.2 2.96-0.76mdk) (from FSF-g77 version 0.5.26 20000731 (Mandrake Linux 8.2 2.96-0.76 mdk)) Driving: g77 -v -c -xf77-version /dev/null -xnone Reading specs from /usr/lib/gcc-lib/i586-mandrake-linux-gnu/2.96/specs gcc version 2.96 20000731 (Mandrake Linux 8.2 2.96-0.76mdk) /usr/lib/gcc-lib/i586-mandrake-linux-gnu/2.96/tradcpp0 -lang-fortran -v -D__GNUC__=2 -D__GNUC_MINOR__=96 -D__GNUC_PATCHLEVEL__ =0 -D__ELF__ -Dunix -Dlinux -D__ELF__ -D__unix__ -D__linux__ -D__unix -D__linux -Asystem(posix) -D__NO_INLINE__ -Acpu(i386) -Am achine(i386) -Di386 -D__i386 -D__i386__ -D__tune_pentium__ /dev/null /dev/null GNU traditional CPP version 2.96 20000731 (Mandrake Linux 8.2 2.96-0.76mdk) /usr/lib/gcc-lib/i586-mandrake-linux-gnu/2.96/f771 -fnull-version -quiet -dumpbase g77-version.f -version -fversion -o /tmp/cc 2P3TbQ.s /dev/null GNU F77 version 2.96 20000731 (Mandrake Linux 8.2 2.96-0.76mdk) (i586-mandrake-linux-gnu) compiled by GNU C version 2.96 200007 31 (Mandrake Linux 8.2 2.96-0.76mdk). GNU Fortran Front End version 0.5.26 20000731 (Mandrake Linux 8.2 2.96-0.76mdk) as -V -Qy -o /tmp/ccPNFOWI.o /tmp/cc2P3TbQ.s GNU assembler version 2.11.92.0.12 (i586-mandrake-linux-gnu) using BFD version 2.11.92.0.12 20011121 ld -m elf_i386 -dynamic-linker /lib/ld-linux.so.2 -o /tmp/cctpS0QJ /tmp/ccPNFOWI.o /usr/lib/gcc-lib/i586-mandrake-linux-gnu/2. 96/../../../crt1.o /usr/lib/gcc-lib/i586-mandrake-linux-gnu/2.96/../../../crti.o /usr/lib/gcc-lib/i586-mandrake-linux-gnu/2.96/ crtbegin.o -L/usr/lib/gcc-lib/i586-mandrake-linux-gnu/2.96 -L/usr/lib/gcc-lib/i586-mandrake-linux-gnu/2.96/../../.. -lg2c -lm - lgcc -lc -lgcc /usr/lib/gcc-lib/i586-mandrake-linux-gnu/2.96/crtend.o /usr/lib/gcc-lib/i586-mandrake-linux-gnu/2.96/../../../cr tn.o /tmp/cctpS0QJ __G77_LIBF77_VERSION__: 0.5.26 20000731 (prerelease) @(#)LIBF77 VERSION 19991115 __G77_LIBI77_VERSION__: 0.5.26 20000731 (prerelease) @(#) LIBI77 VERSION pjw,dmg-mods 19991115 __G77_LIBU77_VERSION__: 0.5.26 20000731 (prerelease) @(#) LIBU77 VERSION 19980709 -------------------------------------------------------------------------- 3) Python version: 2.2 (#1, Feb 24 2002, 16:21:58) [GCC 2.96 20000731 (Mandrake Linux 8.2 2.96-0.76mdk)] -------------------------------------------------------------------------- 4) Python Numeric version: 22.0 -------------------------------------------------------------------------- 5) f2py version: 2.32.225-1419%) -------------------------------------------------------------------------- 6) ATLAS version, the locations of atlas and lapack libraries, building atlas_info: FOUND: libraries = ['lapack', 'f77blas', 'cblas', 'atlas'] library_dirs = ['/usr/lib'] running build_ext building 'atlas_version' extension gcc -DNDEBUG -O3 -fomit-frame-pointer -pipe -mcpu=pentiumpro -march=i586 -ffast-math -fno-strength-reduce -fPIC -I/usr/include/python2.2 -c /usr/lib/python2.2/site-packages/scipy/linalg/atlas_version.c -o build/temp.linux-i686-2.2/atlas_version.o gcc: /usr/lib/python2.2/site-packages/scipy/linalg/atlas_version.c: Aucun fichier ou r?pertoire de ce type gcc: No input files error: command 'gcc' failed with exit status 1 --------------------------------------------------------------------------- [oudet at localhost linalg]$ python -c 'import atlas_version' Traceback (most recent call last): File "", line 1, in ? ImportError: No module named atlas_version (!!!!!!!!!!!!!!!!!!) The rpm file was : atlas-3.4.1-4mdk.noarch.rpm --------------------------------------------------------------------------- 7) The output of the following commands [root at localhost site-packages]# python scipy_distutils/system_info.py atlas_info: FOUND: libraries = ['lapack', 'f77blas', 'cblas', 'atlas'] library_dirs = ['/usr/lib'] blas_info: FOUND: libraries = ['blas'] library_dirs = ['/usr/lib'] blas_src_info: NOT AVAILABLE dfftw_info: NOT AVAILABLE dfftw_threads_info: NOT AVAILABLE djbfft_info: NOT AVAILABLE fftw_info: NOT AVAILABLE fftw_threads_info: NOT AVAILABLE lapack_info: FOUND: libraries = ['lapack'] library_dirs = ['/usr/lib'] lapack_src_info: NOT AVAILABLE sfftw_info: NOT AVAILABLE sfftw_threads_info: NOT AVAILABLE x11_info: FOUND: libraries = ['X11'] library_dirs = ['/usr/X11R6/lib'] include_dirs = ['/usr/X11R6/include'] --------------------------------------------------------------------------- [root at localhost site-packages]# python scipy_distutils/command/build_flib.py Gnu 0.5.26 --------------------------------------------------------------------------- 8) Feel free to add any other relevant information. For example, the The main problem seams to be with the linalg module : >>> scipy.test() No test suite found for scipy.__cvs_version__ creating test suite for: scipy.common !! FAILURE building test for scipy.common :1: SyntaxError: invalid syntax (line 77) (in ?) No test suite found for scipy.helper No test suite found for scipy.helpmod No test suite found for scipy.pilutil No test suite found for scipy.proc No test suite found for scipy.scipy_tempfile No test suite found for scipy.sync No test suite found for scipy.io creating test suite for: scipy.linalg.basic Erreur de segmentation --------------------------------------------------------------------------- Thanks a lot for help, sorry again, ------------------------------------------------------------------- Edouard Oudet Universit? Louis Pasteur 7 rue Ren? Descartes 67000 Strasbourg +33 (0)3 90 24 02 72 (office) +33 (0)3 88 32 31 98 (home) ------------------------------------------------------------------- From pearu at cens.ioc.ee Wed Jan 22 14:06:30 2003 From: pearu at cens.ioc.ee (Pearu Peterson) Date: Wed, 22 Jan 2003 21:06:30 +0200 (EET) Subject: [SciPy-user] Precisions : Install CVS on Mandrake i686 In-Reply-To: Message-ID: On Wed, 22 Jan 2003, Edouard Oudet wrote: > > Here are the precisions that I should give in my first email (sorry)...: > -------------------------------------------------------------------------- > 6) ATLAS version, the locations of atlas and lapack libraries, building > > atlas_info: > FOUND: > libraries = ['lapack', 'f77blas', 'cblas', 'atlas'] > library_dirs = ['/usr/lib'] > > running build_ext > building 'atlas_version' extension > gcc -DNDEBUG -O3 -fomit-frame-pointer -pipe -mcpu=pentiumpro -march=i586 > -ffast-math -fno-strength-reduce -fPIC -I/usr/include/python2.2 -c > /usr/lib/python2.2/site-packages/scipy/linalg/atlas_version.c -o > build/temp.linux-i686-2.2/atlas_version.o > gcc: /usr/lib/python2.2/site-packages/scipy/linalg/atlas_version.c: Aucun > fichier ou r?pertoire de ce type > gcc: No input files > error: command 'gcc' failed with exit status 1 Hmm, how did you exactly tried to build/install scipy? It looks like atlas_version.c (and may be other source files as well) is installed to /usr/lib/python2.2/site-packages/scipy/linalg but it shouldn't. Btw, could you translate the error message "Aucunfichier ou r?pertoire de ce type"? > --------------------------------------------------------------------------- > [oudet at localhost linalg]$ python -c 'import atlas_version' > Traceback (most recent call last): > File "", line 1, in ? > ImportError: No module named atlas_version > > (!!!!!!!!!!!!!!!!!!) See above. > --------------------------------------------------------------------------- > 8) Feel free to add any other relevant information. For example, the > The main problem seams to be with the linalg module : > > >>> scipy.test() > No test suite found for scipy.__cvs_version__ > creating test suite for: scipy.common > !! FAILURE building test for scipy.common > :1: SyntaxError: invalid syntax (line 77) (in ?) Are you sure that you are using the latest CVS version of Scipy? Also, if you have earlier (say, 2-6 months ago) tried to build scipy then remove the following files from linalg/ directory (that probably contain the missing `min` symbol that you reported earlier): cblas.pyf flapack.pyf clapack.pyf fblas.pyf and rebuild scipy from scratch, that is, remove the build/ directory as well all scipy related directories from /usr/lib/python2.2/site-packages/ before hitting python setup.py install in scipy local CVS directory. Pearu From eric at enthought.com Wed Jan 22 18:13:16 2003 From: eric at enthought.com (eric jones) Date: Wed, 22 Jan 2003 17:13:16 -0600 Subject: [SciPy-user] Problem with plt.plot In-Reply-To: Message-ID: <00d801c2c26b$db570e40$8901a8c0@ERICDESKTOP> Yes. Plt is based on wxPython, so it is a required package. The gplt package has a similar interface, but will work on any machine that has gnuplot installed. I think the windows version of scipy even comes with a gnuplot binary. Also, I believe gnuplot comes on most Unix systems. xplt should work on any Unix system also. Regards, eric ---------------------------------------------- eric jones 515 Congress Ave www.enthought.com Suite 1614 512 536-1057 Austin, Tx 78701 > -----Original Message----- > From: scipy-user-admin at scipy.net [mailto:scipy-user-admin at scipy.net] On > Behalf Of Agustin Lobo > Sent: Wednesday, January 22, 2003 10:18 AM > To: scipy-user at scipy.net > Subject: [SciPy-user] Problem with plt.plot > > > Hi! > > I'm following the tutorial > "Minimal Python for Scientific Computing" > > and get to this problem when I try > to use plt: > > >>> from scipy import * > >>> import gui_thread > >> Exception in thread Thread-1: > Traceback (most recent call last): > File "/var/tmp/python-root//usr/lib/python2.1/threading.py", line 378, > in __bootstrap > self.run() > File "/var/tmp/python-root//usr/lib/python2.1/threading.py", line 366, > in run > apply(self.__target, self.__args, self.__kwargs) > File > "/tmp/SciPyTest/linux2/lib/python2.1/site-packages/gui_thread/main.py", > line 38, in gui_thread > File > "/tmp/SciPyTest/linux2/lib/python2.1/site- > packages/gui_thread/gui_thread_guts.py", > line 44, in ? > ImportError: No module named wxPython.wx > > should download and install wxPython just to use plt.plot ? > > Thanks > > Agus > > Dr. Agustin Lobo > Instituto de Ciencias de la Tierra (CSIC) > Lluis Sole Sabaris s/n > 08028 Barcelona SPAIN > tel 34 93409 5410 > fax 34 93411 0012 > alobo at ija.csic.es > > > _______________________________________________ > SciPy-user mailing list > SciPy-user at scipy.net > http://www.scipy.net/mailman/listinfo/scipy-user From alobo at ija.csic.es Thu Jan 23 03:58:57 2003 From: alobo at ija.csic.es (Agustin Lobo) Date: Thu, 23 Jan 2003 09:58:57 +0100 (MET) Subject: [SciPy-user] xplt does not display the graphic Message-ID: Hi, I have this problem: >>> x=arange(0,pi,pi/100) >>> y=sin(x) >>> xplt.plot(x,y) A graphic window opens, but there is no plot in it. What's wrong? Thanks Agus Dr. Agustin Lobo Instituto de Ciencias de la Tierra (CSIC) Lluis Sole Sabaris s/n 08028 Barcelona SPAIN tel 34 93409 5410 fax 34 93411 0012 alobo at ija.csic.es From alobo at ija.csic.es Thu Jan 23 11:14:43 2003 From: alobo at ija.csic.es (Agustin Lobo) Date: Thu, 23 Jan 2003 17:14:43 +0100 (MET) Subject: [SciPy-user] writing a binary file Message-ID: How do I write a float array to a binary file? There is a readFloatArray() in Module Scientific, but not an equivalent writeFloatArray(), while writeArray() writes an ascii file. Thanks Agus Dr. Agustin Lobo Instituto de Ciencias de la Tierra (CSIC) Lluis Sole Sabaris s/n 08028 Barcelona SPAIN tel 34 93409 5410 fax 34 93411 0012 alobo at ija.csic.es From oudet at math.u-strasbg.fr Thu Jan 23 11:49:01 2003 From: oudet at math.u-strasbg.fr (Edouard Oudet) Date: Thu, 23 Jan 2003 17:49:01 +0100 (MET) Subject: [SciPy-user] Precisions : Install CVS on Mandrake i686 In-Reply-To: Message-ID: I succeeded now in installing scipy (the main problem was due to the fact that I didn't use a CVS scipy, shame on me !!). Thanks Pearu for your suggestions.... Nevertheless, there is still a small problem : I can not load scipy with the command >import scipy I get the error : ImportError: No module named fastumath scipy cannot be imported from its source directory. Change to another directory and try again. But from the other hand it is possible to load modules with a ''manual loading'' like import scipy.linalg Does anyone have an idea how is it possible to load all scipy in one step. Thanks a lot, Edouard. ------------------------------------------------------------------- Edouard Oudet Universit? Louis Pasteur 7 rue Ren? Descartes 67000 Strasbourg +33 (0)3 90 24 02 72 (office) +33 (0)3 88 32 31 98 (home) ------------------------------------------------------------------- From pearu at cens.ioc.ee Thu Jan 23 12:11:30 2003 From: pearu at cens.ioc.ee (Pearu Peterson) Date: Thu, 23 Jan 2003 19:11:30 +0200 (EET) Subject: [SciPy-user] Precisions : Install CVS on Mandrake i686 In-Reply-To: Message-ID: On Thu, 23 Jan 2003, Edouard Oudet wrote: > Nevertheless, there is still a small problem : I can not load scipy with > the command > > >import scipy > > I get the error : > > ImportError: No module named fastumath > scipy cannot be imported from its source directory. > Change to another directory and try again. > > But from the other hand it is possible to load modules with a ''manual > loading'' like > > import scipy.linalg Actually, nothing useful happens with this import (for more details see scipy mailing list archives). > Does anyone have an idea how is it possible to load all scipy in one step. Yes, read the error message above and follow what it suggests. Let me know how to improve this message if it wasn't clear enough and not worth following it from the first place. Pearu From oliphant at ee.byu.edu Thu Jan 23 13:57:55 2003 From: oliphant at ee.byu.edu (Travis Oliphant) Date: Thu, 23 Jan 2003 11:57:55 -0700 (MST) Subject: [SciPy-user] xplt does not display the graphic In-Reply-To: Message-ID: > > Hi, > I have this problem: > >>> x=arange(0,pi,pi/100) > >>> y=sin(x) > >>> xplt.plot(x,y) > > A graphic window opens, but there > is no plot in it. What's wrong? We'll need more information. What platform are you on? where do you issue those commands? A terminal (the python interpreter), a script? I use xplt all of the time and so I know it can work. For example, if you try to use xplt from IDLE or another GUI interface, you will probably have difficulties. -teo From oliphant at ee.byu.edu Thu Jan 23 14:05:47 2003 From: oliphant at ee.byu.edu (Travis Oliphant) Date: Thu, 23 Jan 2003 12:05:47 -0700 (MST) Subject: [SciPy-user] writing a binary file In-Reply-To: Message-ID: > > How do I write a float array to a binary file? > There is a > readFloatArray() > in Module Scientific, but not an equivalent > writeFloatArray(), while writeArray() > writes an ascii file. > Look at the io subpackge. There are all kinds of techniques for saving binary data. You can use Pickle to write the Python Object to a file. You can use fid = io.fopen() and fid.write(a) where a is a Numeric array to write to a file. Thus, fid = io.fopen('myfile.dat','w') fid.write(a) fid.close() will write the array a to the file 'myfile.dat' as a linear stream of bytes. This does not store information about the structure of the array, however. It's main purpose is to interface with binary files directly. If you want the data to be stored for a later Python session use Pickle import cPickle cPickle.dump(a, 'myfile.pkl', 1) # write the object to the file (binary) a = cPickle.load('myfile.pkl') # read the object back from the file. This preserves the object. a can be just about any kind of Python object (a dictionary of Numeric arrays, for example). So, if you are only saving data from Python for reading back into Python. cPickle in the standard library is the way to go. The io subpackage in scipy is available for interfacing to other file formats. -teo -- Travis Oliphant Assistant Professor 459 CB Electrical and Computer Engineering Brigham Young University Provo, UT 84602 Tel: (801) 422-3108 oliphant.travis at ieee.org From oliphant at ee.byu.edu Thu Jan 23 14:06:34 2003 From: oliphant at ee.byu.edu (Travis Oliphant) Date: Thu, 23 Jan 2003 12:06:34 -0700 (MST) Subject: [SciPy-user] Precisions : Install CVS on Mandrake i686 In-Reply-To: Message-ID: > > ImportError: No module named fastumath > scipy cannot be imported from its source directory. > Change to another directory and try again. Did you follow this instruction? Don't run scipy from it's installation directory. It will mess up the import system. -teo =========================== Travis Oliphant Assistant Professor 459 CB Electrical and Computer Engineering Brigham Young University Provo, UT 84602 Tel: (801) 422-3108 oliphant.travis at ieee.org From schnoerr at mailzone.com Thu Jan 23 17:41:14 2003 From: schnoerr at mailzone.com (=?ISO-8859-1?Q?Claudius_Schn=F6rr?=) Date: Thu, 23 Jan 2003 23:41:14 +0100 Subject: [SciPy-user] Q: Linear constrained quadratic programming code in python? Message-ID: <3E306F8A.2080505@mailzone.com> Hello, I'm looking for optimization code in python for a problem like this: min( 0.5 x^T Q x + p^T x ) subject to a_i^T x <= b_i, i \in I (inequality constraints) a_i^T x = b_i, i \in E (equality constraints) c_i <= x <= c_s where Q may be a positive definite Matrix, a_i, b_i are vectors, c_i and c_s are bounding vectors, and x is the vector to minimize the problem. Has anybody code for this in python? Please send also a copy to schnoerr at mailzone.com Thank you very much in advance, Claudius From karshi.hasanov at utoronto.ca Thu Jan 23 21:48:32 2003 From: karshi.hasanov at utoronto.ca (Karshi) Date: Thu, 23 Jan 2003 21:48:32 -0500 Subject: [SciPy-user] integrate? Message-ID: <3E30A980.3010300@utoronto.ca> Hi all, I wanna integrate 3d data ( say, A[i,j,k] array), and am wondering if SciPy has a fuction ( or module) for this task. Thanks From oliphant.travis at ieee.org Fri Jan 24 02:24:33 2003 From: oliphant.travis at ieee.org (Travis Oliphant) Date: 24 Jan 2003 00:24:33 -0700 Subject: [SciPy-user] integrate? In-Reply-To: <3E30A980.3010300@utoronto.ca> References: <3E30A980.3010300@utoronto.ca> Message-ID: <1043393074.16573.1.camel@travis.local.net> On Thu, 2003-01-23 at 19:48, Karshi wrote: > Hi all, > > I wanna integrate 3d data ( say, A[i,j,k] array), and am > wondering if SciPy has a fuction ( or module) for this task. > Look at scipy.integrate >>> from scipy import * >>> import scipy >>> info(scipy.integrate) Methods for Integrating Functions given function object. quad -- General purpose integration. dblquad -- General purpose double integration. tplquad -- General purpose triple integration. fixed_quad -- Integrate func(x) using Gaussian quadrature of order n. quadrature -- Integrate with given tolerance using Gaussian quadrature. romberg -- Integrate func using Romberg integration. Methods for Integrating Functions given fixed samples. trapz -- Use trapezoidal rule to compute integral from samples. cumtrapz -- Use trapezoidal rule to cumulatively compute integral. simps -- Use Simpson's rule to compute integral from samples. romb -- Use Romberg Integration to compute integral from (2**k + 1) evenly-spaced samples. See the special module's orthogonal polynomials (special) for Gaussian quadrature roots and weights for other weighting factors and regions. Interface to numerical integrators of ODE systems. odeint -- General integration of ordinary differential equations. ode -- Integrate ODE using vode routine. I presume you want to estimate integrals from samples? There are three methods scipy.integrate.trapz scipy.integrate.simps From falted at openlc.org Fri Jan 24 03:58:19 2003 From: falted at openlc.org (Francesc Alted) Date: Fri, 24 Jan 2003 09:58:19 +0100 Subject: [SciPy-user] writing a binary file In-Reply-To: References: Message-ID: <200301240958.20030.falted@openlc.org> Hola Agus, A Dijous 23 Gener 2003 17:14, Agustin Lobo va escriure: > How do I write a float array to a binary file? > There is a > readFloatArray() > in Module Scientific, but not an equivalent > writeFloatArray(), while writeArray() > writes an ascii file. You may want to have a try to PyTables (http://pytables.sf.net). It will let you to save arbitrary Numeric arrays in binary format and preserving the object properties. One of the advantages that PyTables offers over cPickle is speed. So, if your dataset is large maybe it is worth to have it a try. Adeu, -- Francesc Alted From bryan.cole at teraview.co.uk Fri Jan 24 07:58:15 2003 From: bryan.cole at teraview.co.uk (bryan cole) Date: 24 Jan 2003 12:58:15 +0000 Subject: [SciPy-user] [Q] - reading and writing ASCII files In-Reply-To: <200301240958.20030.falted@openlc.org> References: <200301240958.20030.falted@openlc.org> Message-ID: <1043413095.4921.12.camel@bryan.teraviewhq.local> Hi, I'm feeling a bit stupid today but how do I use the scipy.io.read_array() function? I'm trying to read in a simple tab-delimited two-column ascii dataset into an array with no success. I can write ascii arrays to disk, but reading them in always fails. What am I missing? Here's an example session: [bryan at bryan bryan]$ python Python 2.2.1 (#1, Aug 30 2002, 12:15:30) [GCC 3.2 20020822 (Red Hat Linux Rawhide 3.2-4)] on linux2 Type "help", "copyright", "credits" or "license" for more information. >>> from scipy import * exceptions.ImportError: /usr/lib/python2.2/site-packages/scipy/linalg/clapack.so: undefined symbol: clapack_sgesv >>> x=arange(0,10,0.1) >>> y=cos(x) >>> z=transpose(array([x,y])) >>> f=io.write_array('myarray.dat',z) >>> f.close() >>> #ok my array is now on disk ... >>> f=open('myarray.dat','r+b') >>> w=io.read_array(f) Traceback (most recent call last): File "", line 1, in ? File "/usr/lib/python2.2/site-packages/scipy/io/array_import.py", line 319, in read_array colsize, collist = getcolumns(ascii_object, columns, sep) File "/usr/lib/python2.2/site-packages/scipy/io/array_import.py", line 247, in getcolumns val = process_line(firstline, separator, collist, [Numeric.Float]*N, 0) File "/usr/lib/python2.2/site-packages/scipy/io/array_import.py", line 227, in process_line vals[k] = extract_columns(arlist, collist[k], atype[k], missing) File "/usr/lib/python2.2/site-packages/scipy/io/array_import.py", line 205, in extract_columns return numpyio.convert_objectarray(toconvlist, atype, missing) SystemError: Objects/methodobject.c:33: bad argument to internal function >>> If anyone has an example scripts to illustrate this, I'd appreciate seeing them. cheers, Bryan -- Bryan Cole Teraview Ltd., 302-304 Cambridge Science Park, Milton Road, Cambridge CB4 0WG, United Kingdom. tel: +44 (1223) 435380 / 435386 (direct-dial) fax: +44 (1223) 435382 From bryan.cole at teraview.co.uk Fri Jan 24 09:13:12 2003 From: bryan.cole at teraview.co.uk (bryan cole) Date: 24 Jan 2003 14:13:12 +0000 Subject: [SciPy-user] [Q] - reading and writing ASCII files In-Reply-To: <1043413095.4921.12.camel@bryan.teraviewhq.local> References: <200301240958.20030.falted@openlc.org> <1043413095.4921.12.camel@bryan.teraviewhq.local> Message-ID: <1043417592.4921.14.camel@bryan.teraviewhq.local> PS. I'm using Scipy version '0.2.0_alpha_102.3553' Bryan -- Bryan Cole Teraview Ltd., 302-304 Cambridge Science Park, Milton Road, Cambridge CB4 0WG, United Kingdom. tel: +44 (1223) 435380 / 435386 (direct-dial) fax: +44 (1223) 435382 From alobo at ija.csic.es Fri Jan 24 10:27:55 2003 From: alobo at ija.csic.es (Agustin Lobo) Date: Fri, 24 Jan 2003 16:27:55 +0100 (MET) Subject: [SciPy-user] writing a binary file In-Reply-To: Message-ID: ##oliphant at ee.byu.edu writes### > Look at the io subpackge. There are all kinds of techniques for saving > binary data. How can I learn how to use it? >>> import io >>> dir(io) ['__builtins__', '__doc__', '__file__', '__name__', 'os', 'rpy_input', 'rpy_output', 'rpy_showfiles', 'showfiles_common', 'showfiles_tty', 'sys'] >>> help(io.rpy_input) rpy_input(prompt, n) None Also: >>> help(read) Traceback (most recent call last): File "", line 1, in ? help(read) NameError: name 'read' is not defined or: >>> help(io.write) Traceback (most recent call last): File "", line 1, in ? help(io.write) AttributeError: 'scipy.io' module has no attribute 'write' While >>> help(fid.read) does work and outputs the help information. What's fid? is it a subpackage? It's not listed in io. Otherwise, your suggestion with fid = io.fopen(), fid.read() and fid.write() works fine, thanks a lot. Agus From alobo at ija.csic.es Fri Jan 24 10:39:01 2003 From: alobo at ija.csic.es (Agustin Lobo) Date: Fri, 24 Jan 2003 16:39:01 +0100 (MET) Subject: [SciPy-user] array indexing Message-ID: Why is this? >>> t array([0, 1, 2, 3]) >>> t[0:2] array([0, 1]) Why is not array([0,1,2]) ? Am I not asking for positions 0,1 and 2 in t ? Thanks Dr. Agustin Lobo Instituto de Ciencias de la Tierra (CSIC) Lluis Sole Sabaris s/n 08028 Barcelona SPAIN tel 34 93409 5410 fax 34 93411 0012 alobo at ija.csic.es From yann.ledu at noos.fr Fri Jan 24 13:26:47 2003 From: yann.ledu at noos.fr (Yann Le Du) Date: Fri, 24 Jan 2003 19:26:47 +0100 (CET) Subject: [SciPy-user] array indexing In-Reply-To: Message-ID: On Fri, 24 Jan 2003, Agustin Lobo wrote: > > Why is this? > > >>> t > array([0, 1, 2, 3]) > >>> t[0:2] > array([0, 1]) > > Why is not > array([0,1,2]) ? Am I not asking > for positions 0,1 and 2 in t ? That's the way indexing works in Python : you can understand it by looking at 0 and 2 inside [0:2] as meaning "from separator 0 to separator 2", where the separators are the things that separate the array values from one another : see http://www.python.org/doc/current/tut/node5.html#SECTION005140000000000000000 Yann From alobo at paleo.ija.csic.es Mon Jan 27 08:52:20 2003 From: alobo at paleo.ija.csic.es (Agustin Lobo) Date: Mon, 27 Jan 2003 14:52:20 +0100 (MET) Subject: [SciPy-user] xplt does not display the graphic In-Reply-To: Message-ID: Thanks Travis, Yes the problem was that I was using IDLE. If I run python from the command line, it works. Nevertheles, it still has some odd behaviour. For example, if after displaying a graph I close the Pygist window, I cannot plot anymore: >>> xplt.plot(array([10,11,12]),array([1,2,3])) Traceback (most recent call last): File "", line 1, in ? File "/tmp/SciPyTest/linux2/lib/python2.1/site-packages/scipy/xplt/Mplot.py", line 447, in plot gist.error: No such coordinate system exists in current graphics window. I have to quit python and start over. Why? Is there any solution? Also, is there any other simple graphics system in python that would wotk from idle? (or another gui for linux, actually idle has the problem that you cannot change the default font size, as far as I've seen). Agus Dr. Agustin Lobo Instituto de Ciencias de la Tierra (CSIC) Lluis Sole Sabaris s/n 08028 Barcelona SPAIN tel 34 93409 5410 fax 34 93411 0012 alobo at ija.csic.es On Thu, 23 Jan 2003, Travis Oliphant wrote: > > > > Hi, > > I have this problem: > > >>> x=arange(0,pi,pi/100) > > >>> y=sin(x) > > >>> xplt.plot(x,y) > > > > A graphic window opens, but there > > is no plot in it. What's wrong? > > We'll need more information. What platform are you on? where do you issue > those commands? A terminal (the python interpreter), a script? > > I use xplt all of the time and so I know it can work. > > For example, if you try to use xplt from IDLE or another GUI interface, > you will probably have difficulties. > > -teo > > _______________________________________________ > SciPy-user mailing list > SciPy-user at scipy.net > http://www.scipy.net/mailman/listinfo/scipy-user > From oliphant at ee.byu.edu Mon Jan 27 11:30:14 2003 From: oliphant at ee.byu.edu (Travis Oliphant) Date: Mon, 27 Jan 2003 09:30:14 -0700 (MST) Subject: [SciPy-user] array indexing In-Reply-To: Message-ID: > Why is this? > > >>> t > array([0, 1, 2, 3]) > >>> t[0:2] > array([0, 1]) > > Why is not > array([0,1,2]) ? Am I not asking > for positions 0,1 and 2 in t ? Now, array slicing follows the same syntax as Python itself. Namely, the last index is not included. You are requesting positions 0 and 1 only with your slice notation t[0:2] -Travis O. From oliphant at ee.byu.edu Mon Jan 27 11:43:25 2003 From: oliphant at ee.byu.edu (Travis Oliphant) Date: Mon, 27 Jan 2003 09:43:25 -0700 (MST) Subject: [SciPy-user] xplt does not display the graphic In-Reply-To: Message-ID: > Thanks Travis, > > Yes the problem was that I was using IDLE. > If I run python from the command line, it works. > Nevertheles, it still has some odd behaviour. For example, > if after displaying a graph I close the Pygist window, > I cannot plot anymore: > I thought I fixed this problem. I'm trying to remember what the issue is. Try: xplt.window(0) before plotting for now. I'll get this fixed soon. -teo From alobo at paleo.ija.csic.es Tue Jan 28 06:52:21 2003 From: alobo at paleo.ija.csic.es (Agustin Lobo) Date: Tue, 28 Jan 2003 12:52:21 +0100 (MET) Subject: [SciPy-user] xplt does not display the graphic In-Reply-To: Message-ID: Travis, Maybe it's because I'm still on python2.1. Using xplt.window(0) works fine, thanks. Is there a manual/tutorial somewhere? Or should I use the one for plt/gnplt in http://www.scipy.org/site_content/tutorials/plot_tutorial ? Agus Dr. Agustin Lobo Instituto de Ciencias de la Tierra (CSIC) Lluis Sole Sabaris s/n 08028 Barcelona SPAIN tel 34 93409 5410 fax 34 93411 0012 alobo at ija.csic.es On Mon, 27 Jan 2003, Travis Oliphant wrote: > > Thanks Travis, > > > > Yes the problem was that I was using IDLE. > > If I run python from the command line, it works. > > Nevertheles, it still has some odd behaviour. For example, > > if after displaying a graph I close the Pygist window, > > I cannot plot anymore: > > > > I thought I fixed this problem. I'm trying to remember what the issue is. > > Try: > > xplt.window(0) > > before plotting for now. > > I'll get this fixed soon. > > -teo > > > _______________________________________________ > SciPy-user mailing list > SciPy-user at scipy.net > http://www.scipy.net/mailman/listinfo/scipy-user > From alobo at paleo.ija.csic.es Tue Jan 28 07:23:46 2003 From: alobo at paleo.ija.csic.es (Agustin Lobo) Date: Tue, 28 Jan 2003 13:23:46 +0100 (MET) Subject: [SciPy-user] ploting in scipy Message-ID: 1. I understand that plt requires wxpython, which seems to me too an extensive package for creating graphic applications. All I need is ploting relatively simple graphics, so this seems not to be the way to go. Am I wrong? 2. For gplt, if I do, as suggested in the tutorial: >>> from scipy import gplt >>> gplt.plot((1,2,3)) Traceback (most recent call last): File "", line 1, in ? File "/tmp/SciPyTest/linux2/lib/python2.1/site-packages/scipy/gplt/interface.py", line 106, in plot File "/tmp/SciPyTest/linux2/lib/python2.1/site-packages/scipy/gplt/pyPlot.py", line 132, in plot File "/tmp/SciPyTest/linux2/lib/python2.1/site-packages/scipy/gplt/pyPlot.py", line 702, in _init_plot File "/tmp/SciPyTest/linux2/lib/python2.1/site-packages/scipy/gplt/pyPlot.py", line 820, in _send IOError: [Errno 32] Broken pipe Any idea on what's wrong? Thanks Agus Dr. Agustin Lobo Instituto de Ciencias de la Tierra (CSIC) Lluis Sole Sabaris s/n 08028 Barcelona SPAIN tel 34 93409 5410 fax 34 93411 0012 alobo at ija.csic.es From otttr440 at student.liu.se Tue Jan 28 07:42:08 2003 From: otttr440 at student.liu.se (Otto Tronarp) Date: 28 Jan 2003 13:42:08 +0100 Subject: [SciPy-user] ploting in scipy In-Reply-To: References: Message-ID: <1043757727.1299.11.camel@mathcore2> Verify that you have gnuplot installed on your machine and that it is in the path (what does which gnuplot returns?) Regards, Otto On Tue, 2003-01-28 at 13:23, Agustin Lobo wrote: > 1. I understand that plt requires wxpython, which seems to me > too an extensive package for creating graphic applications. All I need > is ploting relatively simple graphics, so this seems not to be the way to > go. Am I wrong? > > 2. For gplt, if I do, as suggested in the tutorial: > >>> from scipy import gplt > >>> gplt.plot((1,2,3)) > Traceback (most recent call last): > File "", line 1, in ? > File > "/tmp/SciPyTest/linux2/lib/python2.1/site-packages/scipy/gplt/interface.py", > line 106, in plot > File > "/tmp/SciPyTest/linux2/lib/python2.1/site-packages/scipy/gplt/pyPlot.py", > line 132, in plot File > "/tmp/SciPyTest/linux2/lib/python2.1/site-packages/scipy/gplt/pyPlot.py", > line 702, in _init_plot > File > "/tmp/SciPyTest/linux2/lib/python2.1/site-packages/scipy/gplt/pyPlot.py", > line 820, in _send > IOError: [Errno 32] Broken pipe > > Any idea on what's wrong? > > Thanks > > Agus > > Dr. Agustin Lobo > Instituto de Ciencias de la Tierra (CSIC) > Lluis Sole Sabaris s/n > 08028 Barcelona SPAIN > tel 34 93409 5410 > fax 34 93411 0012 > alobo at ija.csic.es > > > > _______________________________________________ > SciPy-user mailing list > SciPy-user at scipy.net > http://www.scipy.net/mailman/listinfo/scipy-user -- Otto Tronarp From alobo at paleo.ija.csic.es Tue Jan 28 08:59:42 2003 From: alobo at paleo.ija.csic.es (Agustin Lobo) Date: Tue, 28 Jan 2003 14:59:42 +0100 (MET) Subject: [SciPy-user] ploting in scipy In-Reply-To: <1043757727.1299.11.camel@mathcore2> Message-ID: Sorry, I should have figured out this one on my own. I removed quite a number of packages from my system to save disk space and gnuplot was among them. I've installed gnuplot again and gplt works fine now. Agus Dr. Agustin Lobo Instituto de Ciencias de la Tierra (CSIC) Lluis Sole Sabaris s/n 08028 Barcelona SPAIN tel 34 93409 5410 fax 34 93411 0012 alobo at ija.csic.es On 28 Jan 2003, Otto Tronarp wrote: > Verify that you have gnuplot installed on your machine and that it is in > the path (what does which gnuplot returns?) > > Regards, > Otto > > On Tue, 2003-01-28 at 13:23, Agustin Lobo wrote: > > 1. I understand that plt requires wxpython, which seems to me > > too an extensive package for creating graphic applications. All I need > > is ploting relatively simple graphics, so this seems not to be the way to > > go. Am I wrong? > > > > 2. For gplt, if I do, as suggested in the tutorial: > > >>> from scipy import gplt > > >>> gplt.plot((1,2,3)) > > Traceback (most recent call last): > > File "", line 1, in ? > > File > > "/tmp/SciPyTest/linux2/lib/python2.1/site-packages/scipy/gplt/interface.py", > > line 106, in plot > > File > > "/tmp/SciPyTest/linux2/lib/python2.1/site-packages/scipy/gplt/pyPlot.py", > > line 132, in plot File > > "/tmp/SciPyTest/linux2/lib/python2.1/site-packages/scipy/gplt/pyPlot.py", > > line 702, in _init_plot > > File > > "/tmp/SciPyTest/linux2/lib/python2.1/site-packages/scipy/gplt/pyPlot.py", > > line 820, in _send > > IOError: [Errno 32] Broken pipe > > > > Any idea on what's wrong? > > > > Thanks > > > > Agus > > > > Dr. Agustin Lobo > > Instituto de Ciencias de la Tierra (CSIC) > > Lluis Sole Sabaris s/n > > 08028 Barcelona SPAIN > > tel 34 93409 5410 > > fax 34 93411 0012 > > alobo at ija.csic.es > > > > > > > > _______________________________________________ > > SciPy-user mailing list > > SciPy-user at scipy.net > > http://www.scipy.net/mailman/listinfo/scipy-user > -- > Otto Tronarp > > _______________________________________________ > SciPy-user mailing list > SciPy-user at scipy.net > http://www.scipy.net/mailman/listinfo/scipy-user > From alobo at paleo.ija.csic.es Tue Jan 28 09:15:23 2003 From: alobo at paleo.ija.csic.es (Agustin Lobo) Date: Tue, 28 Jan 2003 15:15:23 +0100 (MET) Subject: [SciPy-user] shell for scipy Message-ID: In addition to idle, is there any other interactive shell for scipy under linux? Could people recommend? Thanks Agus Dr. Agustin Lobo Instituto de Ciencias de la Tierra (CSIC) Lluis Sole Sabaris s/n 08028 Barcelona SPAIN tel 34 93409 5410 fax 34 93411 0012 alobo at ija.csic.es From otttr440 at student.liu.se Tue Jan 28 09:16:03 2003 From: otttr440 at student.liu.se (Otto Tronarp) Date: 28 Jan 2003 15:16:03 +0100 Subject: [SciPy-user] shell for scipy In-Reply-To: References: Message-ID: <1043763362.1299.27.camel@mathcore2> Perhaps IPython? Haven't used it my self, but I have heard good things about it. http://www-hep.colorado.edu/~fperez/ipython/ Regards, Otto On Tue, 2003-01-28 at 15:15, Agustin Lobo wrote: > In addition to idle, is there any other interactive > shell for scipy under linux? Could people recommend? > > Thanks > > Agus > > Dr. Agustin Lobo > Instituto de Ciencias de la Tierra (CSIC) > Lluis Sole Sabaris s/n > 08028 Barcelona SPAIN > tel 34 93409 5410 > fax 34 93411 0012 > alobo at ija.csic.es > > > _______________________________________________ > SciPy-user mailing list > SciPy-user at scipy.net > http://www.scipy.net/mailman/listinfo/scipy-user -- Otto Tronarp From yann.ledu at noos.fr Tue Jan 28 09:58:42 2003 From: yann.ledu at noos.fr (Yann Le Du) Date: Tue, 28 Jan 2003 15:58:42 +0100 (CET) Subject: [SciPy-user] shell for scipy In-Reply-To: <1043763362.1299.27.camel@mathcore2> Message-ID: On 28 Jan 2003, Otto Tronarp wrote: > Perhaps IPython? Haven't used it my self, but I have heard good things > about it. > http://www-hep.colorado.edu/~fperez/ipython/ Yes, I use it and find it excellent. Y From baecker at physik.tu-dresden.de Tue Jan 28 10:24:50 2003 From: baecker at physik.tu-dresden.de (baecker at physik.tu-dresden.de) Date: Tue, 28 Jan 2003 16:24:50 +0100 (CET) Subject: [SciPy-user] shell for scipy In-Reply-To: References: Message-ID: In the longer run I think it would be nice to have some kind of worksheet environment (like in maple/mathematica/matlab...) which for example also allows to embed the results of plots and has a nice hierarchical help browser. Would something like this for example be possible with Scintilla (eg. under wxPython) ? Are the any plans in this direction ? Arnd On Tue, 28 Jan 2003, Yann Le Du wrote: > On 28 Jan 2003, Otto Tronarp wrote: > > > Perhaps IPython? Haven't used it my self, but I have heard good things > > about it. > > http://www-hep.colorado.edu/~fperez/ipython/ > > Yes, I use it and find it excellent. > > Y > > _______________________________________________ > SciPy-user mailing list > SciPy-user at scipy.net > http://www.scipy.net/mailman/listinfo/scipy-user > From nwagner at mecha.uni-stuttgart.de Wed Jan 29 06:51:41 2003 From: nwagner at mecha.uni-stuttgart.de (Nils Wagner) Date: Wed, 29 Jan 2003 12:51:41 +0100 Subject: [SciPy-user] General_purpose root-finding routines Message-ID: <3E37C04D.DD5CAA85@mecha.uni-stuttgart.de> Hi all, Afaik, thers is only one nonlinear m u l t i-variable equation solver in scipy so far, that is fsolve. Is there someone thinking about an extension with regard to this field ? Nils From alobo at ija.csic.es Wed Jan 29 09:09:30 2003 From: alobo at ija.csic.es (Agustin Lobo) Date: Wed, 29 Jan 2003 15:09:30 +0100 (MET) Subject: [SciPy-user] python/scipy help system Message-ID: I have problems to identify what function I need to perform a given task. For example, someone told me about sys.path.append(). Now, lets assume I don' know about this function, how would I have find it? If I do info(sys), there is some information on path, but does not even mention path.append. Also, there is no information on this function in python/html/lib/module-sys.html Is there any other help system? Thanks Agus Dr. Agustin Lobo Instituto de Ciencias de la Tierra (CSIC) Lluis Sole Sabaris s/n 08028 Barcelona SPAIN tel 34 93409 5410 fax 34 93411 0012 alobo at ija.csic.es From boonstb at cmg.FCNBD.COM Wed Jan 29 10:09:07 2003 From: boonstb at cmg.FCNBD.COM (Brian Boonstra) Date: Wed, 29 Jan 2003 09:09:07 -0600 Subject: [SciPy-user] python/scipy help system In-Reply-To: References: Message-ID: <200301291509.JAA08526@wo1203.cmg.FCNBD.COM> Agustin Lobo wrote: > > Is there any other help system? > I'm not sure it is a whole lot better, but: ------------------------------- from pydoc import help import scipy help(scipy) ------------------------------- From eric at enthought.com Wed Jan 29 10:11:08 2003 From: eric at enthought.com (eric jones) Date: Wed, 29 Jan 2003 09:11:08 -0600 Subject: [SciPy-user] General_purpose root-finding routines In-Reply-To: <3E37C04D.DD5CAA85@mecha.uni-stuttgart.de> Message-ID: <026e01c2c7a8$a9d7cac0$8901a8c0@ERICDESKTOP> I'm all for more solvers/optimizers -- especially a constrained solver for nonlinear optimization problems. The best I know of is cfsqp (Andre Tits is the author I believe), but it isn't open source. If we could get one of comparable quality in SciPy, that would be a *huge* addition. There are probably a lot of more specialized solvers for specific problem types that would also be big additions. I don't know of anyone working on such things. eric ---------------------------------------------- eric jones 515 Congress Ave www.enthought.com Suite 1614 512 536-1057 Austin, Tx 78701 > -----Original Message----- > From: scipy-user-admin at scipy.net [mailto:scipy-user-admin at scipy.net] On > Behalf Of Nils Wagner > Sent: Wednesday, January 29, 2003 5:52 AM > To: scipy-user at scipy.net > Subject: [SciPy-user] General_purpose root-finding routines > > Hi all, > > Afaik, thers is only one nonlinear m u l t i-variable equation solver in > scipy so far, that is fsolve. > Is there someone thinking about an extension with regard to this field ? > > Nils > _______________________________________________ > SciPy-user mailing list > SciPy-user at scipy.net > http://www.scipy.net/mailman/listinfo/scipy-user From boonstb at cmg.FCNBD.COM Wed Jan 29 10:21:11 2003 From: boonstb at cmg.FCNBD.COM (Brian Boonstra) Date: Wed, 29 Jan 2003 09:21:11 -0600 Subject: [SciPy-user] General_purpose root-finding routines In-Reply-To: <026e01c2c7a8$a9d7cac0$8901a8c0@ERICDESKTOP> References: <026e01c2c7a8$a9d7cac0$8901a8c0@ERICDESKTOP> Message-ID: <200301291521.JAA08586@wo1203.cmg.FCNBD.COM> > I'm all for more solvers/optimizers -- especially a constrained solver > for nonlinear optimization problems. The best I know of is cfsqp (Andre > Tits is the author I believe), but it isn't open source. If we could > get one of comparable quality in SciPy, that would be a *huge* addition. I don't have a particular need for this, but I think that MINPACK is quite well regarded. Brian From nwagner at mecha.uni-stuttgart.de Wed Jan 29 10:49:59 2003 From: nwagner at mecha.uni-stuttgart.de (Nils Wagner) Date: Wed, 29 Jan 2003 16:49:59 +0100 Subject: [SciPy-user] General_purpose root-finding routines References: <026e01c2c7a8$a9d7cac0$8901a8c0@ERICDESKTOP> Message-ID: <3E37F827.D5A8E301@mecha.uni-stuttgart.de> eric jones schrieb: > > I'm all for more solvers/optimizers -- especially a constrained solver > for nonlinear optimization problems. The best I know of is cfsqp (Andre > Tits is the author I believe), but it isn't open source. If we could > get one of comparable quality in SciPy, that would be a *huge* addition. > > There are probably a lot of more specialized solvers for specific > problem types that would also be big additions. > > I don't know of anyone working on such things. > > eric BTW, I wonder if I can solve the following problem with scipy P(\lambda(\mu), \mu) x(\mu) = 0 0.5 (x^H x -1) = 0 (Normalization condition) y^T \frac{\partial P}{\partial \lambda} x = 0 P is a so-called lambda-matrix, for example P = \lambda^2 M + \lambda D + K + \mu S x right-eigenvector y left-eigenvector ^H : hermitian ^T : transpose \mu : real parameter \lambda : eigenvalue A small example is appreciated Thanks in advance Nils > > ---------------------------------------------- > eric jones 515 Congress Ave > www.enthought.com Suite 1614 > 512 536-1057 Austin, Tx 78701 > > > -----Original Message----- > > From: scipy-user-admin at scipy.net [mailto:scipy-user-admin at scipy.net] > On > > Behalf Of Nils Wagner > > Sent: Wednesday, January 29, 2003 5:52 AM > > To: scipy-user at scipy.net > > Subject: [SciPy-user] General_purpose root-finding routines > > > > Hi all, > > > > Afaik, thers is only one nonlinear m u l t i-variable equation solver > in > > scipy so far, that is fsolve. > > Is there someone thinking about an extension with regard to this field > ? > > > > Nils > > _______________________________________________ > > SciPy-user mailing list > > SciPy-user at scipy.net > > http://www.scipy.net/mailman/listinfo/scipy-user > > _______________________________________________ > SciPy-user mailing list > SciPy-user at scipy.net > http://www.scipy.net/mailman/listinfo/scipy-user From h.jansen at fel.tno.nl Wed Jan 29 11:44:49 2003 From: h.jansen at fel.tno.nl (H Jansen) Date: Wed, 29 Jan 2003 17:44:49 +0100 Subject: [SciPy-user] General_purpose root-finding routines References: <026e01c2c7a8$a9d7cac0$8901a8c0@ERICDESKTOP> Message-ID: <3E380501.964A2568@fel.tno.nl> So am I! I've been working on a nontrivial optimal control problem over the last year, and missing a constraint nonlinear solver in Python, it has been a pain to do this all in C++. I've posted for its need within a Python environment before but is seems it's still on the wanted list. Instead, I've been using the Omuses/HQP SQP (sequential quadratic problem, sparse matrix) solver which is available on Sourceforge (http://hqp.sourceforge.net and http://www.systemtechnik.tu-ilmenau.de/~fg_opt/omuses/hqp.html) and is free under GPL Version 2 (the only constraint nonlinear solver that is free, as far as I know) --- I would propose it as a candidate for Python! A few aspects must be mentioned: 1. It has already a tcl interface for controlling the solution procedure (parameters and order of execution) --- it should be replace by Python ASAP; 2. Too bad, it lacks a scripting interface for problem specification ... (for how long ...?) 3. The optimization problem (the relation between independent (i.e. control) and dependent variables) must be specified with a special type of double, so-called adouble (=active double, provided by the free ADOL-C library): it provides "on the fly" gradient information of the problem at hand. Not underestimating the work that should be done, I do believe that it's a viable candidate to be interfaced with Python! Henk Jansen eric jones wrote: > > I'm all for more solvers/optimizers -- especially a constrained solver > for nonlinear optimization problems. The best I know of is cfsqp (Andre > Tits is the author I believe), but it isn't open source. If we could > get one of comparable quality in SciPy, that would be a *huge* addition. > > There are probably a lot of more specialized solvers for specific > problem types that would also be big additions. > > I don't know of anyone working on such things. > > eric > > ---------------------------------------------- > eric jones 515 Congress Ave > www.enthought.com Suite 1614 > 512 536-1057 Austin, Tx 78701 > > > -----Original Message----- > > From: scipy-user-admin at scipy.net [mailto:scipy-user-admin at scipy.net] > On > > Behalf Of Nils Wagner > > Sent: Wednesday, January 29, 2003 5:52 AM > > To: scipy-user at scipy.net > > Subject: [SciPy-user] General_purpose root-finding routines > > > > Hi all, > > > > Afaik, thers is only one nonlinear m u l t i-variable equation solver > in > > scipy so far, that is fsolve. > > Is there someone thinking about an extension with regard to this field > ? > > > > Nils > > _______________________________________________ > > SciPy-user mailing list > > SciPy-user at scipy.net > > http://www.scipy.net/mailman/listinfo/scipy-user > > _______________________________________________ > SciPy-user mailing list > SciPy-user at scipy.net > http://www.scipy.net/mailman/listinfo/scipy-user -------------- next part -------------- A non-text attachment was scrubbed... Name: h.jansen.vcf Type: text/x-vcard Size: 479 bytes Desc: Card for H Jansen URL: From alobo at ija.csic.es Thu Jan 30 06:33:01 2003 From: alobo at ija.csic.es (Agustin Lobo) Date: Thu, 30 Jan 2003 12:33:01 +0100 (MET) Subject: [SciPy-user] reshape Message-ID: Hi! I have 3D array (time, latitude, longitude) and want to convert it into another (latitude, longitude, time). I've tried reshape(), but just changes dimensions, not actually the structure of the array. How should I do it? Also, how could I have got the answer by myself using the help system? (give a man a fish..., give a man a fishing rod) Thanks Agus Dr. Agustin Lobo Instituto de Ciencias de la Tierra (CSIC) Lluis Sole Sabaris s/n 08028 Barcelona SPAIN tel 34 93409 5410 fax 34 93411 0012 alobo at ija.csic.es From alobo at ija.csic.es Thu Jan 30 07:01:31 2003 From: alobo at ija.csic.es (Agustin Lobo) Date: Thu, 30 Jan 2003 13:01:31 +0100 (MET) Subject: [SciPy-user] reload Message-ID: What's the best way to modify function with the editor and then get this function actualized in the python shell? This is what I'm doing: 1. import leo_ncd3D 2. from leo_ncd3D import * 3. I make a change in function leo_ncd3Dlin() which is within file leo_ncd3D.py. 4. reload(leo_ncd3D) 5. leo_ncd3D.leo_ncd3Dlin() If in (4) I do leo_ncd3Dlin() instead of leo_ncd3D.leo_ncd3Dlin(), I still use the older version. Is this the normal way? (I'm using idle, which I don't know if it's the best shell) Thanks! Agus Dr. Agustin Lobo Instituto de Ciencias de la Tierra (CSIC) Lluis Sole Sabaris s/n 08028 Barcelona SPAIN tel 34 93409 5410 fax 34 93411 0012 alobo at ija.csic.es From bryan.cole at teraview.co.uk Thu Jan 30 07:27:10 2003 From: bryan.cole at teraview.co.uk (bryan cole) Date: 30 Jan 2003 12:27:10 +0000 Subject: [SciPy-user] reshape In-Reply-To: References: Message-ID: <1043929630.19796.16.camel@bryan.teraviewhq.local> try: transpose(myarray, (1,2,0)) Have you got the Numeric documentation? SciPy is built in top of Numeric, so all Numeric functions are found in SciPy (AFAIK). The Numeric documentation (at least the pdf-version) is somewhat frustrating because the functions are not listed in any particular order as far as I can tell (alphabetical order would have been an improvement!). I can only suggest you browse the Numeric docs extensively! Once you know what function you need, help(func) will bring up the details. Bryan On Thu, 2003-01-30 at 11:33, Agustin Lobo wrote: > Hi! > > I have 3D array (time, latitude, longitude) and > want to convert it into another (latitude, longitude, time). > I've tried reshape(), but just changes dimensions, not actually > the structure of the array. How should I do it? > Also, how could I have got the answer by myself using the > help system? (give a man a fish..., give a man a fishing rod) > > Thanks > > Agus > > > Dr. Agustin Lobo > Instituto de Ciencias de la Tierra (CSIC) > Lluis Sole Sabaris s/n > 08028 Barcelona SPAIN > tel 34 93409 5410 > fax 34 93411 0012 > alobo at ija.csic.es > > > _______________________________________________ > SciPy-user mailing list > SciPy-user at scipy.net > http://www.scipy.net/mailman/listinfo/scipy-user -- Bryan Cole Teraview Ltd., 302-304 Cambridge Science Park, Milton Road, Cambridge CB4 0WG, United Kingdom. tel: +44 (1223) 435380 / 435386 (direct-dial) fax: +44 (1223) 435382 From nwagner at mecha.uni-stuttgart.de Thu Jan 30 07:13:09 2003 From: nwagner at mecha.uni-stuttgart.de (Nils Wagner) Date: Thu, 30 Jan 2003 13:13:09 +0100 Subject: [SciPy-user] LOCA 1.1 Library of continuation algorithms Message-ID: <3E3916D5.383086B4@mecha.uni-stuttgart.de> Hi all, LOCA is a new software library for performing bifurcation analysis of large-scale applications. Is it possible to include this library in scipy ? http://www.cs.sandia.gov/LOCA/ for further details Nils From oudet at irmasrv2.u-strasbg.fr Thu Jan 30 07:42:05 2003 From: oudet at irmasrv2.u-strasbg.fr (Edouard Oudet) Date: Thu, 30 Jan 2003 13:42:05 +0100 (CET) Subject: [SciPy-user] Problem with gui_thread ? Message-ID: As reported by Nils wagner (see http://scipy.net/pipermail/scipy-user/2002-November/000932.html), I didn't succeed in ploting (with plt.plot) when I run python with the -i option (I have the same error). I should underline, that exactly the same instructions worked perfectly when I run python first and then gave the same instructions (I have the same problem when I run a script with the function execfile('...')). Does anyone have a solution ? Thanks for help, Edouard. ------------------------------------------------------------------- Edouard Oudet Universit? Louis Pasteur 7 rue Ren? Descartes 67000 Strasbourg +33 (0)3 90 24 02 72 (office) +33 (0)3 88 32 31 98 (home) ------------------------------------------------------------------- From jsteve17 at tampabay.rr.com Thu Jan 30 13:38:25 2003 From: jsteve17 at tampabay.rr.com (Jeffrey Stephens) Date: Thu, 30 Jan 2003 13:38:25 -0500 Subject: [SciPy-user] Installation Problem Message-ID: <200301301338.25711.jsteve17@tampabay.rr.com> I just finished installing SciPy-0.2.0_alpha144.4350. I run the following: $python Python 2.2.2 (#1, Nov 22 2002, 17:25:34) [GCC 2.96 20000731 (Red Hat Linux 7.3 2.96-112)] on linux2 Type "help", "copyright", "credits" or "license" for more information. >>> import scipy >>> scipy.test(level=6) No test suite found for scipy.__cvs_version__ creating test suite for: scipy.common !! FAILURE building test for scipy.common :1: SyntaxError: invalid syntax (test_common.py, line 77) (in ?) No test suite found for scipy.helper No test suite found for scipy.helpmod No test suite found for scipy.pilutil No test suite found for scipy.proc No test suite found for scipy.scipy_tempfile No test suite found for scipy.sync No test suite found for scipy.io creating test suite for: scipy.linalg.basic Segmentation fault I get this error regardless of what level the test is set to. Does anyone know what is wrong with my install? Running RH 7.3, kernel 2.4.18-19.7.x, KDE 3.0.3, gcc 2.96-113, Regards, Jeff Stephens From pearu at cens.ioc.ee Thu Jan 30 13:49:49 2003 From: pearu at cens.ioc.ee (Pearu Peterson) Date: Thu, 30 Jan 2003 20:49:49 +0200 (EET) Subject: [SciPy-user] Installation Problem In-Reply-To: <200301301338.25711.jsteve17@tampabay.rr.com> Message-ID: On Thu, 30 Jan 2003, Jeffrey Stephens wrote: > I just finished installing SciPy-0.2.0_alpha144.4350. I run the following: > > $python > Python 2.2.2 (#1, Nov 22 2002, 17:25:34) > [GCC 2.96 20000731 (Red Hat Linux 7.3 2.96-112)] on linux2 > Type "help", "copyright", "credits" or "license" for more information. > >>> import scipy > >>> scipy.test(level=6) > No test suite found for scipy.__cvs_version__ > creating test suite for: scipy.common > !! FAILURE building test for scipy.common > :1: SyntaxError: invalid syntax (test_common.py, line 77) (in ?) > No test suite found for scipy.helper > No test suite found for scipy.helpmod > No test suite found for scipy.pilutil > No test suite found for scipy.proc > No test suite found for scipy.scipy_tempfile > No test suite found for scipy.sync > No test suite found for scipy.io > creating test suite for: scipy.linalg.basic > Segmentation fault > > I get this error regardless of what level the test is set to. Does anyone > know what is wrong with my install? > > Running RH 7.3, kernel 2.4.18-19.7.x, KDE 3.0.3, gcc 2.96-113, What Numeric version are you using? Did you upgrade it recently? If yes, make sure that you installed Numeric with --force flag, e.g. python setup.py install --force Also, I suggest using CVS version of SciPy. Pearu From jsteve17 at tampabay.rr.com Thu Jan 30 14:20:29 2003 From: jsteve17 at tampabay.rr.com (Jeffrey Stephens) Date: Thu, 30 Jan 2003 14:20:29 -0500 Subject: [SciPy-user] Re: Installation Problem Message-ID: <200301301420.29534.jsteve17@tampabay.rr.com> Using Numeric-22.0. I built a Numeric RPM from source and installed w/o any command line options. Should I rebuild/reinstall? I'll download SciPy from CVS and try this first. Thanks. Regards, Jeff S > What Numeric version are you using? Did you upgrade it recently? > If yes, make sure that you installed Numeric with --force flag, e.g. > python setup.py install --force > Also, I suggest using CVS version of SciPy. > > Pearu From sugino at brandeis.edu Thu Jan 30 09:51:11 2003 From: sugino at brandeis.edu (Ken Sugino) Date: Thu, 30 Jan 2003 14:51:11 +0000 Subject: [SciPy-user] reload In-Reply-To: References: Message-ID: <20030130145111.6e34c424.sugino@brandeis.edu> If you do : >>> from leo_ncd3D import * after step 4, you can use 'leo_ncd3Dlin()' as newly defined function. For instance objects, you have to assign __class__ to reloaded class: >>> import foo # module foo >>> a = foo.bar() # bar class in foo >>> reload(foo) >>> a.__class__ == foo.bar 0 # different class object >>> a.__class__ = foo.bar # have to reassign to use new class definition (like methods, class attributes) There is a recipe for autoreloading in ASPN: http://aspn.activestate.com/ASPN/Cookbook/Python/Recipe/160164 Cheers, ken On Thu, 30 Jan 2003 13:01:31 +0100 (MET) Agustin Lobo wrote: > > What's the best way to modify function with the editor > and then get this function actualized in the python > shell? > This is what I'm doing: > > 1. import leo_ncd3D > 2. from leo_ncd3D import * > 3. I make a change in function leo_ncd3Dlin() > which is within file leo_ncd3D.py. > 4. reload(leo_ncd3D) > 5. leo_ncd3D.leo_ncd3Dlin() > > If in (4) I do leo_ncd3Dlin() instead of > leo_ncd3D.leo_ncd3Dlin(), I still use the older version. > > Is this the normal way? > (I'm using idle, which I don't know if it's the > best shell) > > Thanks! > > Agus > > Dr. Agustin Lobo > Instituto de Ciencias de la Tierra (CSIC) > Lluis Sole Sabaris s/n > 08028 Barcelona SPAIN > tel 34 93409 5410 > fax 34 93411 0012 > alobo at ija.csic.es > > > _______________________________________________ > SciPy-user mailing list > SciPy-user at scipy.net > http://www.scipy.net/mailman/listinfo/scipy-user From jsteve17 at tampabay.rr.com Thu Jan 30 15:07:13 2003 From: jsteve17 at tampabay.rr.com (Jeffrey Stephens) Date: Thu, 30 Jan 2003 15:07:13 -0500 Subject: [SciPy-user] Re: Installation Problem - Test Results In-Reply-To: <200301301420.29534.jsteve17@tampabay.rr.com> References: <200301301420.29534.jsteve17@tampabay.rr.com> Message-ID: <200301301507.13665.jsteve17@tampabay.rr.com> After I removed the previous SciPy and installed the CVS version the problem was corrected; however, I am getting 'some' errors when I run the test import scipy scipy.test(level=1) I have attached the test results in a separate file. I wonder if any of the failures have to do with the fact that my blas (3.0-14) and lapack (3.0-14) libraries were installed by using the stock RPMS supplied with RedHat 7.3. Should I upgrade to the most recent and complete versions or does this make a difference vis-a-vis the errors I got? Regards, Jeff S On Thursday 30 January 2003 02:20 pm, Jeffrey Stephens wrote: > Using Numeric-22.0. I built a Numeric RPM from source and installed w/o > any command line options. Should I rebuild/reinstall? I'll download SciPy > from CVS and try this first. Thanks. > > Regards, > Jeff S > > > What Numeric version are you using? Did you upgrade it recently? > > If yes, make sure that you installed Numeric with --force flag, e.g. > > python setup.py install --force > > Also, I suggest using CVS version of SciPy. > > > > Pearu > > _______________________________________________ > SciPy-user mailing list > SciPy-user at scipy.net > http://www.scipy.net/mailman/listinfo/scipy-user -------------- next part -------------- $ python Python 2.2.2 (#1, Nov 22 2002, 17:25:34) [GCC 2.96 20000731 (Red Hat Linux 7.3 2.96-112)] on linux2 Type "help", "copyright", "credits" or "license" for more information. >>> import scipy >>> scipy.test(level=1) No test suite found for scipy.__cvs_version__ creating test suite for: scipy.common No test suite found for scipy.helper No test suite found for scipy.helpmod No test suite found for scipy.pilutil No test suite found for scipy.proc No test suite found for scipy.scipy_tempfile No test suite found for scipy.sync No test suite found for scipy.io creating test suite for: scipy.linalg.basic creating test suite for: scipy.linalg.blas **************************************************************** WARNING: Importing fblas failed with the following exception: ----------- exceptions.ImportError: /usr/lib/python2.2/site-packages/scipy/linalg/fblas.so: undefined symbol: srotmg_ ----------- See scipy/INSTALL.txt for troubleshooting. **************************************************************** **************************************************************** WARNING: Importing cblas failed with the following exception: ----------- exceptions.ImportError: cannot import name cblas ----------- See scipy/INSTALL.txt for troubleshooting. Note that if atlas library is not found by scipy/system_info.py, then scipy skips building cblas and uses fblas instead. **************************************************************** creating test suite for: scipy.linalg.decomp No test suite found for scipy.linalg.flinalg No test suite found for scipy.linalg.interface_gen creating test suite for: scipy.linalg.lapack **************************************************************** WARNING: Importing clapack failed with the following exception: ----------- exceptions.ImportError: cannot import name clapack ----------- See scipy/INSTALL.txt for troubleshooting. Notes: * If atlas library is not found by scipy/system_info.py, then scipy skips building clapack and uses flapack instead. **************************************************************** creating test suite for: scipy.linalg.matfuncs !! FAILURE building test for scipy.linalg.matfuncs :1: ImportError: No module named test_matfuncs (in ?) No test suite found for scipy.special.gendoc No test suite found for scipy_base.__cvs_version__ creating test suite for: scipy_base.function_base creating test suite for: scipy_base.index_tricks creating test suite for: scipy_base.limits creating test suite for: scipy_base.matrix_base No test suite found for scipy_base.polynomial No test suite found for scipy_base.scimath No test suite found for scipy_base.scipy_base_version creating test suite for: scipy_base.shape_base creating test suite for: scipy_base.type_check creating test suite for: scipy.special.special No test suite found for scipy.signal creating test suite for: scipy.stats.distributions creating test suite for: scipy.stats.morestats No test suite found for scipy.stats.pstat No test suite found for scipy.stats.rv creating test suite for: scipy.stats.stats No test suite found for scipy.interpolate No test suite found for scipy.integrate No test suite found for anneal No test suite found for scipy.optimize.common_routines No test suite found for scipy.optimize.minpack No test suite found for scipy.optimize.optimize creating test suite for: scipy.optimize.zeros No test suite found for scipy.cluster No test suite found for scipy.cow No test suite found for scipy.ga creating test suite for: scipy.fftpack.basic creating test suite for: scipy.fftpack.helper creating test suite for: scipy.fftpack.pseudo_diffs ......................................................................EEEE.................................................................................................................... Gamma overflow error ...................................................................................................................Testing uniform .Testing norm .Testing lognorm .Testing expon .Testing beta .Testing powerlaw .Testing bradford .Testing burr .Testing fisk .Testing cauchy .Testing halfcauchy .Testing foldcauchy .Testing gamma .Testing gengamma .Testing loggamma .Testing alpha .Testing anglit .Testing arcsine .Testing betaprime .Testing erlang .Testing dgamma .Testing exponweib .Testing exponpow .Testing frechet_l .Testing frechet_r .Testing gilbrat .Testing f .Testing ncf .Testing chi2 .Testing chi .Testing nakagami .Testing genpareto .Testing genextreme .Testing genhalflogistic .Testing pareto .Testing lomax .Testing halfnorm .Testing halflogistic .Testing fatiguelife .Testing foldnorm .Testing ncx2 .Testing t .Testing nct .Testing weibull_min .Testing weibull_max .Testing dweibull .Testing maxwell .Testing rayleigh .Testing genlogistic .Testing logistic .Testing gumbel_l .Testing gumbel_r .Testing gompertz .Testing hypsecant .Testing laplace .Testing reciprocal .Testing triang .Testing tukeylambda ................Ties preclude use of exact statistic. ..Ties preclude use of exact statistic. ..............................................................................................TESTING CONVERGENCE zero should be 1 function f2 cc.bisect : 1.0000000000001952 cc.ridder : 1.0000000000004658 cc.brenth : 1.0000000000002278 cc.brentq : 1.0000000000007097 function f3 cc.bisect : 1.0000000000001952 cc.ridder : 1.0000000000000000 cc.brenth : 0.9999999999999943 cc.brentq : 1.0000000000000000 function f4 cc.bisect : 1.0000000000001952 cc.ridder : 1.0000000000001450 cc.brenth : 0.9999999999997333 cc.brentq : 0.9999999999997725 function f5 cc.bisect : 1.0000000000001952 cc.ridder : 1.0000000000004285 cc.brenth : 1.0000000000014235 cc.brentq : 1.0000000000008802 function f6 cc.bisect : 1.0000000000001952 cc.ridder : 1.0000000000001068 cc.brenth : 1.0000000000002371 cc.brentq : 1.0000000000004587 ............................................ ====================================================================== ERROR: check_random (test_decomp.test_qr) ---------------------------------------------------------------------- Traceback (most recent call last): File "/usr/lib/python2.2/site-packages/scipy/linalg/tests/test_decomp.py", line 295, in check_random q,r = qr(a) File "/usr/lib/python2.2/site-packages/scipy/linalg/decomp.py", line 359, in qr gemm, = get_blas_funcs(('gemm',),(qr,)) File "/usr/lib/python2.2/site-packages/scipy/linalg/blas.py", line 43, in get_blas_funcs if ordering and fblas.has_column_major_storage(arrays[ordering[0][1]]): File "/usr/lib/python2.2/site-packages/scipy_distutils/misc_util.py", line 44, in __getattr__ raise self._info[0],self._info[1] ImportError: /usr/lib/python2.2/site-packages/scipy/linalg/fblas.so: undefined symbol: srotmg_ ====================================================================== ERROR: check_random_complex (test_decomp.test_qr) ---------------------------------------------------------------------- Traceback (most recent call last): File "/usr/lib/python2.2/site-packages/scipy/linalg/tests/test_decomp.py", line 302, in check_random_complex q,r = qr(a) File "/usr/lib/python2.2/site-packages/scipy/linalg/decomp.py", line 359, in qr gemm, = get_blas_funcs(('gemm',),(qr,)) File "/usr/lib/python2.2/site-packages/scipy/linalg/blas.py", line 43, in get_blas_funcs if ordering and fblas.has_column_major_storage(arrays[ordering[0][1]]): File "/usr/lib/python2.2/site-packages/scipy_distutils/misc_util.py", line 44, in __getattr__ raise self._info[0],self._info[1] ImportError: /usr/lib/python2.2/site-packages/scipy/linalg/fblas.so: undefined symbol: srotmg_ ====================================================================== ERROR: check_simple (test_decomp.test_qr) ---------------------------------------------------------------------- Traceback (most recent call last): File "/usr/lib/python2.2/site-packages/scipy/linalg/tests/test_decomp.py", line 283, in check_simple q,r = qr(a) File "/usr/lib/python2.2/site-packages/scipy/linalg/decomp.py", line 359, in qr gemm, = get_blas_funcs(('gemm',),(qr,)) File "/usr/lib/python2.2/site-packages/scipy/linalg/blas.py", line 43, in get_blas_funcs if ordering and fblas.has_column_major_storage(arrays[ordering[0][1]]): File "/usr/lib/python2.2/site-packages/scipy_distutils/misc_util.py", line 44, in __getattr__ raise self._info[0],self._info[1] ImportError: /usr/lib/python2.2/site-packages/scipy/linalg/fblas.so: undefined symbol: srotmg_ ====================================================================== ERROR: check_simple_complex (test_decomp.test_qr) ---------------------------------------------------------------------- Traceback (most recent call last): File "/usr/lib/python2.2/site-packages/scipy/linalg/tests/test_decomp.py", line 288, in check_simple_complex q,r = qr(a) File "/usr/lib/python2.2/site-packages/scipy/linalg/decomp.py", line 359, in qr gemm, = get_blas_funcs(('gemm',),(qr,)) File "/usr/lib/python2.2/site-packages/scipy/linalg/blas.py", line 43, in get_blas_funcs if ordering and fblas.has_column_major_storage(arrays[ordering[0][1]]): File "/usr/lib/python2.2/site-packages/scipy_distutils/misc_util.py", line 44, in __getattr__ raise self._info[0],self._info[1] ImportError: /usr/lib/python2.2/site-packages/scipy/linalg/fblas.so: undefined symbol: srotmg_ ---------------------------------------------------------------------- Ran 518 tests in 21.019s FAILED (errors=4) From fperez at pizero.colorado.edu Thu Jan 30 15:30:06 2003 From: fperez at pizero.colorado.edu (Fernando Perez) Date: Thu, 30 Jan 2003 13:30:06 -0700 (MST) Subject: [SciPy-user] shell for scipy In-Reply-To: Message-ID: On Tue, 28 Jan 2003 baecker at physik.tu-dresden.de wrote: > In the longer run I think it would be nice > to have some kind of worksheet environment > (like in maple/mathematica/matlab...) which for example > also allows to embed the results of plots and > has a nice hierarchical help browser. > Would something like this for example be possible with > Scintilla (eg. under wxPython) ? > Are the any plans in this direction ? > > Arnd > > On Tue, 28 Jan 2003, Yann Le Du wrote: > > > On 28 Jan 2003, Otto Tronarp wrote: > > > > > Perhaps IPython? Haven't used it my self, but I have heard good things > > > about it. > > > http://www-hep.colorado.edu/~fperez/ipython/ > > > > Yes, I use it and find it excellent. Thanks :) Being the main ipython developer, I figured I might want to pitch in a few words. This message happens to coincide with me just having released ipython 0.2.14 after 50 -pre releases. I haven't announced it anywhere, as I've been super busy this week. But the above-referenced url has the files ready for download. I think IPython is indeed quite nice, and in combination with a solid text editor (Xemacs, vi, etc) I find it far better than idle. But of course, I'm biased :) However, for serious work I do think there are some things in ipython which indeed make it better than the others: - tight access to the underlying OS for shell-like operations. This is very handy when doing data manipulation. - the @pdb command to pop up pdb 'inside' crashed code. This often is enough for debugging and in my experience better/easier than stepping with a debugger or putting in the proverbial print statements. - the @run command, both for fast prototyping of modules (no reload issues) and for profiling (-p gives direct profiler control). - the embedding features for tricky debugging situations (think of the .stop/.continue commands in IDL). - the concept of profiles, which allows ipython to adapt to multiple scenarios, even modifying its processing of the command-line syntax. However, I also acknowledge that a graphical notebook-like interface would be a great thing to have in many cases. My mental model of the ideal pyhton environment is Mathematica (even though their interface does have its rough edges, I love the basic idea). 'math' opens a basic command-line interface, while 'mathematica' gives you the notebook-based gui. That's exactly what I'd like to have for python: ipython as the command-line shell, and a gui which uses ipython as its interaction engine, but which can also save the whole session, edit across lines, embed graphics and change fonts, etc. Such a project is fairly ambitious, but I'd like to pitch it here to see if there is community response. Mathematica, maple, matlab & IDL all have graphical desktops and for scipy to dethrone the last two of them, it will probably need one too (I say the last two because I don't see scipy as competing with Mathematica, but definitely as a better environment than matlab or IDL). Currently ipython is fairly solid and stable, but in need of a big internal cleanup. I'm very hesitant to do much to it right now, because if the idea of coordinating with a graphical shell does get support from the scipy community, then the project will become larger. In that case, the redesign of the current internals should be done with an eye on making it easy to drop ipython into any graphical shell (the mythical scipy graphical shell, idle, the new Eric editor, whatever...) I would therefore like to hear from the scipy users/developers about interest on this topic. I honestly think that ipython provides some of the most solid low-level interface of all the current python shells, but it's text only. I definitely can't do the work of writing a large graphical layer above it, but I'd be more than happy to coordinate witha larger team for the necessary rewrites. If there is no interest, ipython will probably continue to slowly improve (it's quite good already and for text-only use I have very few things I see as needed improvements). But without larger community support I doubt I will be able to devote any major amount of time to it. Best regards, Fernando. From pearu at cens.ioc.ee Thu Jan 30 15:30:36 2003 From: pearu at cens.ioc.ee (Pearu Peterson) Date: Thu, 30 Jan 2003 22:30:36 +0200 (EET) Subject: [SciPy-user] Re: Installation Problem - Test Results In-Reply-To: <200301301507.13665.jsteve17@tampabay.rr.com> Message-ID: On Thu, 30 Jan 2003, Jeffrey Stephens wrote: > After I removed the previous SciPy and installed the CVS version the problem > was corrected; however, I am getting 'some' errors when I run the test > > import scipy > scipy.test(level=1) > > I have attached the test results in a separate file. I wonder if any of the > failures have to do with the fact that my blas (3.0-14) and lapack (3.0-14) > libraries were installed by using the stock RPMS supplied with RedHat 7.3. > Should I upgrade to the most recent and complete versions or does this > make a difference vis-a-vis the errors I got? These rpm's provide incomplete blas libraries (that are probably build using blas from lapack tree and this is known to miss certain blas functions that are present in official version of blas, see netlib). I don't know if recent versions of RedHat blas/lapack rpm's contain complete blas but a better solution would be installing atlas instead. See scipy/INSTALL.txt for instructions. Pearu From fperez at pizero.colorado.edu Thu Jan 30 15:33:15 2003 From: fperez at pizero.colorado.edu (Fernando Perez) Date: Thu, 30 Jan 2003 13:33:15 -0700 (MST) Subject: [SciPy-user] reload In-Reply-To: Message-ID: On Thu, 30 Jan 2003, Agustin Lobo wrote: > > What's the best way to modify function with the editor > and then get this function actualized in the python > shell? > This is what I'm doing: > > 1. import leo_ncd3D > 2. from leo_ncd3D import * > 3. I make a change in function leo_ncd3Dlin() > which is within file leo_ncd3D.py. > 4. reload(leo_ncd3D) > 5. leo_ncd3D.leo_ncd3Dlin() > > If in (4) I do leo_ncd3Dlin() instead of > leo_ncd3D.leo_ncd3Dlin(), I still use the older version. > > Is this the normal way? > (I'm using idle, which I don't know if it's the > best shell) You may want to try out ipython (http://www-hep.colorado.edu/~fperez/ipython/). Instead of doing 'import', you can cd to the definition directory and type 'run leo_ncd3d'. This executes the file in the current namespace, so any changes made to the source take effect immediately. That's typically my approach, and for most single-level modules it works well. If you have a complex package with nested structure, ipython offers a 'dreload' version of reload which helps doing recursive reloads. Best regards, Fernando. From answer at tnoo.net Fri Jan 31 01:22:40 2003 From: answer at tnoo.net (=?iso-8859-1?q?Martin_L=FCthi?=) Date: 30 Jan 2003 21:22:40 -0900 Subject: [SciPy-user] Re: reload References: <20030130145111.6e34c424.sugino@brandeis.edu> Message-ID: Ipython http://www-hep.colorado.edu/~fperez/ipython/ helps a lot, especially the automatic jump into the debugger is very convenient. Have fun! Martin -- Martin L?thi answer at tnoo.net From Kasper.Souren at ircam.fr Thu Jan 30 16:20:53 2003 From: Kasper.Souren at ircam.fr (Kasper Souren) Date: Thu, 30 Jan 2003 22:20:53 +0100 (CET) Subject: [SciPy-user] ipython => TeXmacs! In-Reply-To: References: Message-ID: <33145.62.210.226.47.1043961653.squirrel@www.ircam.fr> > I would therefore like to hear from the scipy users/developers about > interest on this topic. I honestly think that ipython provides some of the > most solid low-level interface of all the current python shells, but it's > text only. I definitely can't do the work of writing a large graphical > layer above it, but I'd be more than happy to coordinate witha larger team > for the necessary rewrites. Maybe you should check out TeXmacs. It's a really nice WYSIWYG TeX, it's GPL and it already has interfaces with some shells (Maxima, Pari, Scheme, Shell, Yacas and recently Octave). bye, Kasper From rossini at blindglobe.net Thu Jan 30 17:35:29 2003 From: rossini at blindglobe.net (A.J. Rossini) Date: Thu, 30 Jan 2003 14:35:29 -0800 Subject: [SciPy-user] shell for scipy In-Reply-To: (Fernando Perez's message of "Thu, 30 Jan 2003 13:30:06 -0700 (MST)") References: Message-ID: <87lm12xo3y.fsf@jeeves.blindglobe.net> >>>>> "FP" == Fernando Perez writes: FP> That's exactly what I'd like to have for python: ipython as the command-line FP> shell, and a gui which uses ipython as its interaction engine, but which can FP> also save the whole session, edit across lines, embed graphics and change FP> fonts, etc. FP> Such a project is fairly ambitious, but I'd like to pitch it here to see if FP> there is community response. Mathematica, maple, matlab & IDL all have FP> graphical desktops and for scipy to dethrone the last two of them, it will FP> probably need one too (I say the last two because I don't see scipy as FP> competing with Mathematica, but definitely as a better environment than matlab FP> or IDL). You might look at the Sweave tools in R, combined with ESS (Emacs Speaks Statistics) for interactive construction and evaluation. It isn't interactive in the Mathematica sense, but it is the next best thing -- construction of documents using Noweb/literate programming techniques, and evaluation of "chunks" interactively, or via a "report generation" mechanism. It's not exactly what is being discussed, but is sort of an approximation. best, -tony -- A.J. Rossini Rsrch. Asst. Prof. of Biostatistics U. of Washington Biostatistics rossini at u.washington.edu FHCRC/SCHARP/HIV Vaccine Trials Net rossini at scharp.org -------------- http://software.biostat.washington.edu/ ---------------- FHCRC: M: 206-667-7025 (fax=4812)|Voicemail is pretty sketchy/use Email UW: Th: 206-543-1044 (fax=3286)|Change last 4 digits of phone to FAX (my tuesday/wednesday/friday locations are completely unpredictable.) From fperez at pizero.colorado.edu Thu Jan 30 20:55:01 2003 From: fperez at pizero.colorado.edu (Fernando Perez) Date: Thu, 30 Jan 2003 18:55:01 -0700 (MST) Subject: [SciPy-user] ipython => TeXmacs! In-Reply-To: <33145.62.210.226.47.1043961653.squirrel@www.ircam.fr> Message-ID: On Thu, 30 Jan 2003, Kasper Souren wrote: > > I would therefore like to hear from the scipy users/developers about > > interest on this topic. I honestly think that ipython provides some of the > > most solid low-level interface of all the current python shells, but it's > > text only. I definitely can't do the work of writing a large graphical > > layer above it, but I'd be more than happy to coordinate witha larger team > > for the necessary rewrites. > > Maybe you should check out TeXmacs. It's a really nice WYSIWYG TeX, it's > GPL and it already has interfaces with some shells (Maxima, Pari, Scheme, > Shell, Yacas and recently Octave). Thanks for the feedback. My main thrust would be something in coordination with the scipy project as a whole. They seem to have settled on wxwindows for gui development and pycrust is part of wxwindows, so that might be the most logical starting point. But if there is not enough interest from the scipy crowd, any other option is welcome. However, as I said before, I don't intend to embark on a gui project myself. For my needs, ipython + xemacs is basically perfect. However, if there is enough interest in a gui scientific pyhton shell, I'd be glad to help along with the internal redesign and I can probably write the 'plumbing' necessary for ipython to plug into another external system. But gui programming just doesn't interest me very much, sorry. Cheers, f From DavidA at ActiveState.com Thu Jan 30 23:47:19 2003 From: DavidA at ActiveState.com (David Ascher) Date: Thu, 30 Jan 2003 20:47:19 -0800 Subject: [SciPy-user] ipython => TeXmacs! References: Message-ID: <3E39FFD7.5040108@ActiveState.com> Fernando Perez wrote: >However, as I said before, I don't intend to embark on a gui project myself. >For my needs, ipython + xemacs is basically perfect. However, if there is >enough interest in a gui scientific pyhton shell, I'd be glad to help along >with the internal redesign and I can probably write the 'plumbing' necessary >for ipython to plug into another external system. But gui programming just >doesn't interest me very much, sorry. > > I'm interested, not so much in the scipy-ipython integration (although that would be neat), but in finding a way of using ipython in Komodo, when we get to do our interactive shells. I'm not going to have time to work on that for a while, but I'd love to learn more about ipython and its architecture -- specifically what API you'd like to see between the GUI handling part and the "model" of the shell. --david ascher From bryan.cole at teraview.co.uk Fri Jan 31 05:42:33 2003 From: bryan.cole at teraview.co.uk (bryan cole) Date: 31 Jan 2003 10:42:33 +0000 Subject: [SciPy-user] shell for scipy In-Reply-To: References: Message-ID: <1044009753.21775.33.camel@bryan.teraviewhq.local> I use pycrust + scipy for interactive numerical work and I've been thinking about how this combination could be improved: What I think would be useful would be the ability to execute blocks of code (say ~10 lines or so) at once. Right now, the various python shells force you to execute code line-by-line. I would like to be able to craft data-analysis code in small blocks then when I'm happy, press an "execute" button to run them, see the resulting output then either move onto the next segment or re-edit the current one and re-run it. Each segment would be automatically appended to a python-file on execution, so I could "restore" to any previous point by restarting the interpreter and re-executing the entire lot. There would be a sort of "script history" function (like the command-history function, but with code-blocks instead of single lines). In general I prefer to do data-analysis from scripts, rather than interactively at the command-line. The problem is at re-running the python-process each time to test things out is slow because Scipy takes so long to load each time (OK I'm impatient). Really, it would be better just to keep the interpreter running between code-executions. When I've got time, I'll try and implement something along these lines using the wxPython. I'm not sure about embedding plots though. I'd be happy to hear any comments/ideas on the above. Bryan On Thu, 2003-01-30 at 20:30, Fernando Perez wrote: > On Tue, 28 Jan 2003 baecker at physik.tu-dresden.de wrote: > > > In the longer run I think it would be nice > > to have some kind of worksheet environment > > (like in maple/mathematica/matlab...) which for example > > also allows to embed the results of plots and > > has a nice hierarchical help browser. > > Would something like this for example be possible with > > Scintilla (eg. under wxPython) ? > > Are the any plans in this direction ? > > > > Arnd > > > > On Tue, 28 Jan 2003, Yann Le Du wrote: > > > > > On 28 Jan 2003, Otto Tronarp wrote: > > > > > > > Perhaps IPython? Haven't used it my self, but I have heard good things > > > > about it. > > > > http://www-hep.colorado.edu/~fperez/ipython/ > > > > > > Yes, I use it and find it excellent. > > Thanks :) Being the main ipython developer, I figured I might want to pitch > in a few words. This message happens to coincide with me just having released > ipython 0.2.14 after 50 -pre releases. I haven't announced it anywhere, as > I've been super busy this week. But the above-referenced url has the files > ready for download. > > I think IPython is indeed quite nice, and in combination with a solid text > editor (Xemacs, vi, etc) I find it far better than idle. But of course, I'm > biased :) However, for serious work I do think there are some things in > ipython which indeed make it better than the others: > > - tight access to the underlying OS for shell-like operations. This is very > handy when doing data manipulation. > > - the @pdb command to pop up pdb 'inside' crashed code. This often is enough > for debugging and in my experience better/easier than stepping with a debugger > or putting in the proverbial print statements. > > - the @run command, both for fast prototyping of modules (no reload issues) > and for profiling (-p gives direct profiler control). > > - the embedding features for tricky debugging situations (think of the > .stop/.continue commands in IDL). > > - the concept of profiles, which allows ipython to adapt to multiple > scenarios, even modifying its processing of the command-line syntax. > > However, I also acknowledge that a graphical notebook-like interface would be > a great thing to have in many cases. My mental model of the ideal pyhton > environment is Mathematica (even though their interface does have its rough > edges, I love the basic idea). 'math' opens a basic command-line interface, > while 'mathematica' gives you the notebook-based gui. > > That's exactly what I'd like to have for python: ipython as the command-line > shell, and a gui which uses ipython as its interaction engine, but which can > also save the whole session, edit across lines, embed graphics and change > fonts, etc. > > Such a project is fairly ambitious, but I'd like to pitch it here to see if > there is community response. Mathematica, maple, matlab & IDL all have > graphical desktops and for scipy to dethrone the last two of them, it will > probably need one too (I say the last two because I don't see scipy as > competing with Mathematica, but definitely as a better environment than matlab > or IDL). > > Currently ipython is fairly solid and stable, but in need of a big internal > cleanup. I'm very hesitant to do much to it right now, because if the idea of > coordinating with a graphical shell does get support from the scipy community, > then the project will become larger. In that case, the redesign of the > current internals should be done with an eye on making it easy to drop ipython > into any graphical shell (the mythical scipy graphical shell, idle, the new > Eric editor, whatever...) > > I would therefore like to hear from the scipy users/developers about interest > on this topic. I honestly think that ipython provides some of the most solid > low-level interface of all the current python shells, but it's text only. I > definitely can't do the work of writing a large graphical layer above it, but > I'd be more than happy to coordinate witha larger team for the necessary > rewrites. > > If there is no interest, ipython will probably continue to slowly improve > (it's quite good already and for text-only use I have very few things I see as > needed improvements). But without larger community support I doubt I will be > able to devote any major amount of time to it. > > Best regards, > > Fernando. > > _______________________________________________ > SciPy-user mailing list > SciPy-user at scipy.net > http://www.scipy.net/mailman/listinfo/scipy-user -- Bryan Cole Teraview Ltd., 302-304 Cambridge Science Park, Milton Road, Cambridge CB4 0WG, United Kingdom. tel: +44 (1223) 435380 / 435386 (direct-dial) fax: +44 (1223) 435382 From bryan.cole at teraview.co.uk Fri Jan 31 06:21:52 2003 From: bryan.cole at teraview.co.uk (bryan cole) Date: 31 Jan 2003 11:21:52 +0000 Subject: [SciPy-user] shell for scipy In-Reply-To: <1044009753.21775.33.camel@bryan.teraviewhq.local> References: <1044009753.21775.33.camel@bryan.teraviewhq.local> Message-ID: <1044012112.21780.37.camel@bryan.teraviewhq.local> OK, I've taken a closer look at ipython and I see I can acheive 'code block' editing using the @edit magic function. Looks like I'm now an ipython convert :) Bryan On Fri, 2003-01-31 at 10:42, bryan cole wrote: > I use pycrust + scipy for interactive numerical work and I've been > thinking about how this combination could be improved: > > What I think would be useful would be the ability to execute blocks of > code (say ~10 lines or so) at once. Right now, the various python shells > force you to execute code line-by-line. > > I would like to be able to craft data-analysis code in small blocks then > when I'm happy, press an "execute" button to run them, see the resulting > output then either move onto the next segment or re-edit the current one > and re-run it. Each segment would be automatically appended to a > python-file on execution, so I could "restore" to any previous point by > restarting the interpreter and re-executing the entire lot. There would > be a sort of "script history" function (like the command-history > function, but with code-blocks instead of single lines). > > In general I prefer to do data-analysis from scripts, rather than > interactively at the command-line. The problem is at re-running the > python-process each time to test things out is slow because Scipy takes > so long to load each time (OK I'm impatient). Really, it would be better > just to keep the interpreter running between code-executions. > > When I've got time, I'll try and implement something along these lines > using the wxPython. I'm not sure about embedding plots though. > > I'd be happy to hear any comments/ideas on the above. > > Bryan > > On Thu, 2003-01-30 at 20:30, Fernando Perez wrote: > > On Tue, 28 Jan 2003 baecker at physik.tu-dresden.de wrote: > > > > > In the longer run I think it would be nice > > > to have some kind of worksheet environment > > > (like in maple/mathematica/matlab...) which for example > > > also allows to embed the results of plots and > > > has a nice hierarchical help browser. > > > Would something like this for example be possible with > > > Scintilla (eg. under wxPython) ? > > > Are the any plans in this direction ? > > > > > > Arnd > > > > > > On Tue, 28 Jan 2003, Yann Le Du wrote: > > > > > > > On 28 Jan 2003, Otto Tronarp wrote: > > > > > > > > > Perhaps IPython? Haven't used it my self, but I have heard good things > > > > > about it. > > > > > http://www-hep.colorado.edu/~fperez/ipython/ > > > > > > > > Yes, I use it and find it excellent. > > > > Thanks :) Being the main ipython developer, I figured I might want to pitch > > in a few words. This message happens to coincide with me just having released > > ipython 0.2.14 after 50 -pre releases. I haven't announced it anywhere, as > > I've been super busy this week. But the above-referenced url has the files > > ready for download. > > > > I think IPython is indeed quite nice, and in combination with a solid text > > editor (Xemacs, vi, etc) I find it far better than idle. But of course, I'm > > biased :) However, for serious work I do think there are some things in > > ipython which indeed make it better than the others: > > > > - tight access to the underlying OS for shell-like operations. This is very > > handy when doing data manipulation. > > > > - the @pdb command to pop up pdb 'inside' crashed code. This often is enough > > for debugging and in my experience better/easier than stepping with a debugger > > or putting in the proverbial print statements. > > > > - the @run command, both for fast prototyping of modules (no reload issues) > > and for profiling (-p gives direct profiler control). > > > > - the embedding features for tricky debugging situations (think of the > > .stop/.continue commands in IDL). > > > > - the concept of profiles, which allows ipython to adapt to multiple > > scenarios, even modifying its processing of the command-line syntax. > > > > However, I also acknowledge that a graphical notebook-like interface would be > > a great thing to have in many cases. My mental model of the ideal pyhton > > environment is Mathematica (even though their interface does have its rough > > edges, I love the basic idea). 'math' opens a basic command-line interface, > > while 'mathematica' gives you the notebook-based gui. > > > > That's exactly what I'd like to have for python: ipython as the command-line > > shell, and a gui which uses ipython as its interaction engine, but which can > > also save the whole session, edit across lines, embed graphics and change > > fonts, etc. > > > > Such a project is fairly ambitious, but I'd like to pitch it here to see if > > there is community response. Mathematica, maple, matlab & IDL all have > > graphical desktops and for scipy to dethrone the last two of them, it will > > probably need one too (I say the last two because I don't see scipy as > > competing with Mathematica, but definitely as a better environment than matlab > > or IDL). > > > > Currently ipython is fairly solid and stable, but in need of a big internal > > cleanup. I'm very hesitant to do much to it right now, because if the idea of > > coordinating with a graphical shell does get support from the scipy community, > > then the project will become larger. In that case, the redesign of the > > current internals should be done with an eye on making it easy to drop ipython > > into any graphical shell (the mythical scipy graphical shell, idle, the new > > Eric editor, whatever...) > > > > I would therefore like to hear from the scipy users/developers about interest > > on this topic. I honestly think that ipython provides some of the most solid > > low-level interface of all the current python shells, but it's text only. I > > definitely can't do the work of writing a large graphical layer above it, but > > I'd be more than happy to coordinate witha larger team for the necessary > > rewrites. > > > > If there is no interest, ipython will probably continue to slowly improve > > (it's quite good already and for text-only use I have very few things I see as > > needed improvements). But without larger community support I doubt I will be > > able to devote any major amount of time to it. > > > > Best regards, > > > > Fernando. > > > > _______________________________________________ > > SciPy-user mailing list > > SciPy-user at scipy.net > > http://www.scipy.net/mailman/listinfo/scipy-user -- Bryan Cole Teraview Ltd., 302-304 Cambridge Science Park, Milton Road, Cambridge CB4 0WG, United Kingdom. tel: +44 (1223) 435380 / 435386 (direct-dial) fax: +44 (1223) 435382 From p.berkes at biologie.hu-berlin.de Fri Jan 31 06:26:40 2003 From: p.berkes at biologie.hu-berlin.de (Pietro Berkes) Date: Fri, 31 Jan 2003 12:26:40 +0100 Subject: [SciPy-user] reload In-Reply-To: References: Message-ID: <200301311226.40868.p.berkes@biologie.hu-berlin.de> On Thursday 30 January 2003 13:01, Agustin Lobo wrote: > What's the best way to modify function with the editor > and then get this function actualized in the python > shell? > This is what I'm doing: > > 1. import leo_ncd3D > 2. from leo_ncd3D import * this is the dangerous part: you copy all the functions and variables in the 'leo_ncd3D' module in the current namespace. This means that if you reload the module, you do not modify the copied names (as you notice below). If you want to reload the module (and in general), you should avoid doing something like (2). If you want to avoid typing the whole module name, try this 1. import leo_ncd3D as leo and the use 'leo' as the new module name, or write a small function that explicitly copies the names from the module to the current namespace: def copy_namespace(module, names): global_namespace = globals() for name in names: global_namespace[name] = getattr(module,name, None) and call it after (1) and (4). (This last approach seems ugly to me. I've never tested the function above - it might be wrong.). Note that this approach to copy all names won't work: def copy_namespace(module): globals().update(module.__dict__) This function is going to replace some important variables, such as the name of the current module. > 3. I make a change in function leo_ncd3Dlin() > which is within file leo_ncd3D.py. > 4. reload(leo_ncd3D) > 5. leo_ncd3D.leo_ncd3Dlin() > > If in (4) I do leo_ncd3Dlin() instead of > leo_ncd3D.leo_ncd3Dlin(), I still use the older version. > > Is this the normal way? > (I'm using idle, which I don't know if it's the > best shell) > > Thanks! > > Agus From alobo at ija.csic.es Fri Jan 31 06:46:00 2003 From: alobo at ija.csic.es (Agustin Lobo) Date: Fri, 31 Jan 2003 12:46:00 +0100 (MET) Subject: [SciPy-user] shell for scipy In-Reply-To: <1044009753.21775.33.camel@bryan.teraviewhq.local> Message-ID: On 31 Jan 2003, bryan cole wrote: > I use pycrust + scipy for interactive numerical work and I've been > thinking about how this combination could be improved: > The problem with pycrust is that, as far as I understand, it requires wxpython. Having to install a whole package for developping guis just to have one gui does not make too much sense. Agus From prabhu at aero.iitm.ernet.in Fri Jan 31 08:58:24 2003 From: prabhu at aero.iitm.ernet.in (Prabhu Ramachandran) Date: Fri, 31 Jan 2003 19:28:24 +0530 Subject: [SciPy-user] shell for scipy In-Reply-To: References: Message-ID: <15930.33024.963771.82916@monster.linux.in> >>>>> "FP" == Fernando Perez writes: FP> I think IPython is indeed quite nice, and in combination with IPython is really nice and a must use for any serious Pythoner. :) [snip] FP> I would therefore like to hear from the scipy users/developers FP> about interest on this topic. I honestly think that ipython FP> provides some of the most solid low-level interface of all the FP> current python shells, but it's text only. I definitely can't FP> do the work of writing a large graphical layer above it, but FP> I'd be more than happy to coordinate witha larger team for the FP> necessary rewrites. Well, you know my state of affairs well enough to know that I cant volunteer time but I think the idea of a redesign is good and worth it in the long run. If there is serious interest the rewrite/redesign will only make it easier for those who wish to see a graphical frontend. Besides, if its easy enough perhaps the existing GUI shells can all incorporate the IPython "engine". However, I have no idea if that would be possible at all. In any case I think it would be really nice if IPython could be embedded in any GUI shell. That would really make for an extremely powerful pycrust or pyshell. Good luck! cheers, prabhu From eric at enthought.com Fri Jan 31 18:20:44 2003 From: eric at enthought.com (eric jones) Date: Fri, 31 Jan 2003 17:20:44 -0600 Subject: [SciPy-user] shell for scipy In-Reply-To: <1044009753.21775.33.camel@bryan.teraviewhq.local> Message-ID: <000001c2c97f$64546fc0$8901a8c0@ERICDESKTOP> > The problem is at re-running the > python-process each time to test things out is slow because Scipy takes > so long to load each time (OK I'm impatient). Agreed. It would be cool if there was some way of doing on-demand loading of the sub-modules without explicitly requiring people to import them. I guess this would have to be a python level feature instead of doing it in scipy. If this were possible, it would ameliorate the problem markedly. The effort involved (or even if it is possible -- should be) is not clear to me. eric From sugino at brandeis.edu Fri Jan 31 15:20:04 2003 From: sugino at brandeis.edu (Ken Sugino) Date: Fri, 31 Jan 2003 20:20:04 +0000 Subject: [SciPy-user] shell for scipy In-Reply-To: <000001c2c97f$64546fc0$8901a8c0@ERICDESKTOP> References: <1044009753.21775.33.camel@bryan.teraviewhq.local> <000001c2c97f$64546fc0$8901a8c0@ERICDESKTOP> Message-ID: <20030131202004.53c075c5.sugino@brandeis.edu> Pmw (Python Mega Widget) has a on-demand-loader which may be a good example to start with. http://pmw.sourceforge.net/doc/dynamicloader.html /usr/lib/python2.2/site-packages/Pmw/__init__.py /usr/lib/python2.2/site-packages/Pmw/Pmw_1_1/lib/PmwLoader.py Ken On Fri, 31 Jan 2003 17:20:44 -0600 "eric jones" wrote: > > The problem is at re-running the > > python-process each time to test things out is slow because Scipy > takes > > so long to load each time (OK I'm impatient). > > Agreed. It would be cool if there was some way of doing on-demand > loading of the sub-modules without explicitly requiring people to import > them. I guess this would have to be a python level feature instead of > doing it in scipy. If this were possible, it would ameliorate the > problem markedly. The effort involved (or even if it is possible -- > should be) is not clear to me. > > eric > > > _______________________________________________ > SciPy-user mailing list > SciPy-user at scipy.net > http://www.scipy.net/mailman/listinfo/scipy-user