From pajer at iname.com Sat Jan 3 06:59:50 2004 From: pajer at iname.com (Gary Pajer) Date: Sat, 3 Jan 2004 06:59:50 -0500 Subject: [SciPy-user] compiling on Windows: getting closer References: <3feda8e1a6ae16.34642054@mercury.sabren.com> Message-ID: <001e01c3d1f1$175b78a0$01fd5644@playroom> My hobby of getting scipy to compile on Windows is getting closer to completion. (full report when it's finished) Briefly: I have Enthought Edition 2.3 and ATLAS and LAPACK compiled from current sources (libraries merged) I compiled ATLAS and LAPACK in the cygwin bash shell using the lastest cygwin compiler. I'm compiling scipy in a DOS shell using the Mingw32 compiler that comes with the Enthought Edition using setup_no_chaco.py (that's a separate problem :) ) THe build chugs along for quite some time until it fails while trying to make the specfun library with error: --------------------------------------------------------------------------- Lib\special\specfun\specfun.f:1513: Internal compiler error in emit_swap_insn, a t reg-stack.c:990 Please submit a full bug report, with preprocessed source if appropriate. See for instructions. error: failure during compile (exit status = 1) --------------------------------------------------------------------------- any clues? -gary From pearu at scipy.org Sat Jan 3 09:35:26 2004 From: pearu at scipy.org (Pearu Peterson) Date: Sat, 3 Jan 2004 08:35:26 -0600 (CST) Subject: [SciPy-user] compiling on Windows: getting closer In-Reply-To: <001e01c3d1f1$175b78a0$01fd5644@playroom> Message-ID: On Sat, 3 Jan 2004, Gary Pajer wrote: > My hobby of getting scipy to compile on Windows is getting closer to > completion. > (full report when it's finished) > > Briefly: I have Enthought Edition 2.3 and ATLAS and LAPACK compiled from > current sources (libraries merged) > I compiled ATLAS and LAPACK in the cygwin bash shell using the lastest > cygwin compiler. > > I'm compiling scipy in a DOS shell using the Mingw32 compiler that comes > with the Enthought Edition using setup_no_chaco.py (that's a separate > problem :) ) > > THe build chugs along for quite some time until it fails while trying to > make the specfun library with error: > > --------------------------------------------------------------------------- > Lib\special\specfun\specfun.f:1513: Internal compiler error in > emit_swap_insn, a > t reg-stack.c:990 > Please submit a full bug report, > with preprocessed source if appropriate. > See for instructions. > error: failure during compile (exit status = 1) > --------------------------------------------------------------------------- > > any clues? Could you try to find a work around to this compiler bug? Or just comment the offending function in specfun.f out in order to complete scipy completion. Pearu From pajer at iname.com Sat Jan 3 10:23:19 2004 From: pajer at iname.com (Gary Pajer) Date: Sat, 3 Jan 2004 10:23:19 -0500 Subject: [SciPy-user] compiling on Windows: getting closer References: Message-ID: <000f01c3d20d$84518a50$01fd5644@playroom> > On Sat, 3 Jan 2004, Gary Pajer wrote: > > > My hobby of getting scipy to compile on Windows is getting closer to > > completion. > > (full report when it's finished) > > > > Briefly: I have Enthought Edition 2.3 and ATLAS and LAPACK compiled from > > current sources (libraries merged) > > I compiled ATLAS and LAPACK in the cygwin bash shell using the lastest > > cygwin compiler. > > > > I'm compiling scipy in a DOS shell using the Mingw32 compiler that comes > > with the Enthought Edition using setup_no_chaco.py (that's a separate > > problem :) ) > > > > THe build chugs along for quite some time until it fails while trying to > > make the specfun library with error: > > > > -------------------------------------------------------------------------- - > > Lib\special\specfun\specfun.f:1513: Internal compiler error in > > emit_swap_insn, a > > t reg-stack.c:990 > > Please submit a full bug report, > > with preprocessed source if appropriate. > > See for instructions. > > error: failure during compile (exit status = 1) > > -------------------------------------------------------------------------- - > > > > any clues? > > Could you try to find a work around to this compiler bug? > Or just comment the offending function in specfun.f out in > order to complete scipy completion. > > Pearu OK, I'll do that. This is really a compiler bug? I'm an amateur and hence wasn't sure exactly what the message meant. What compiler do the pros use? I can try a newer mingw32, or perhaps the cygwin gcc. I'll play around. Thanks, -gary From pearu at scipy.org Sat Jan 3 10:55:02 2004 From: pearu at scipy.org (Pearu Peterson) Date: Sat, 3 Jan 2004 09:55:02 -0600 (CST) Subject: [SciPy-user] compiling on Windows: getting closer In-Reply-To: <000f01c3d20d$84518a50$01fd5644@playroom> Message-ID: On Sat, 3 Jan 2004, Gary Pajer wrote: > > > Lib\special\specfun\specfun.f:1513: Internal compiler error in > > > emit_swap_insn, a > > > t reg-stack.c:990 > > > Please submit a full bug report, > > > with preprocessed source if appropriate. > > > See for instructions. > > > error: failure during compile (exit status = 1) > > > > -------------------------------------------------------------------------- > - > > > > > > any clues? > > > > Could you try to find a work around to this compiler bug? > > Or just comment the offending function in specfun.f out in > > order to complete scipy completion. > > > > Pearu > > OK, I'll do that. > This is really a compiler bug? Yes, any 'Internal compiler error' indicates a bug in compiler. google emit_swap_insn gives more info. > I'm an amateur and hence wasn't sure exactly what the message meant. > What compiler do the pros use? > I can try a newer mingw32, or perhaps the cygwin gcc. I'll play around. I am not on Windows (yet), but on Linux compiling specfun.f with gcc has never caused any trouble to me, starting from gcc version 2.95. Google indicates that the compiler bug above is platform dependent. Pearu From scipy at zunzun.com Sat Jan 3 15:53:11 2004 From: scipy at zunzun.com (James R. Phillips) Date: Sat, 3 Jan 2004 15:53:11 -500 Subject: [SciPy-user] Simple least-squares curve fitting example Message-ID: <3ff72bb7205bd1.36540171@mercury.sabren.com> Here is a linear least squares curvefitting example for the distribution if you would like to use it. James Phillips http://zunzun.com import scipy.linalg.basic xdata = [5.357, 5.457, 5.797, 5.936, 6.161, 6.697, 6.731, 6.775, 8.442, 9.769, 9.861] ydata = [0.376, 0.489, 0.874, 1.049, 1.327, 2.054, 2.077, 2.138, 4.744, 7.068, 7.104] matrix = [] for x in xdata: matrix.append([1.0, x, x*x]) # for y = a + bx + cx^2 coeffs = scipy.linalg.basic.lstsq(matrix, ydata)[0] print "scipy.linalg.basic.lstsq curve fitting example" print "fitting data to quadratic equation y = a + bx + cx^2" print "yields: x data y data calc value error" for i in range(len(xdata)): ycalc = coeffs[0] + coeffs[1] * xdata[i] + coeffs[2] * xdata[i] * xdata[i] error = ycalc - ydata[i] print " % .3f % .3f % .3f % .3f" % (xdata[i], ydata[i], ycalc, error) print From pajer at iname.com Sat Jan 3 18:49:54 2004 From: pajer at iname.com (Gary Pajer) Date: Sat, 3 Jan 2004 18:49:54 -0500 Subject: [SciPy-user] compiling on Windows: getting closer References: Message-ID: <002101c3d254$4b350e80$01fd5644@playroom> > Could you try to find a work around to this compiler bug? > Or just comment the offending function in specfun.f out in > order to complete scipy completion. > > Pearu Compiling on Windows, continued. I short-circuited the functions in specfun.f that were causing problems. Those functions are RMN2L and ASWFB, the radial and angular prolate and oblate spherical functions of the first kind. Hopefully, I've not created a land mine. Now specfun.f compiles to completion. But I have a new problem, one that looks odd to my eye. See below. Is there a command called "g" ? BTW, if anyone can tell me that they've built scipy on Windows, and how they did it, please contact me. How did the Enthought people do it? -gary ---------------------------------------------------------------------------- ----------------- c:\mingw\bin\gcc.exe -O2 -w -Wstrict-prototypes -IC:\PYTHON23\include -IC:\P YTHO N23\PC -c Lib\cluster\src/vq_wrap.cpp -o build\temp.win32-2.3\Release\lib\cluste r\src\vq_wrap.o g -shared build\temp.win32-2.3\Release\lib\cluster\src\vq_wrap.o -LC:\PYTHON23\l ibs -LC:\PYTHON23\PCBuild -Lbuild\temp.win32-2.3 -lpython23 -lrootfind -lc_m isc -lcephes -o build\lib.win32-2.3\scipy\cluster\_vq.pyd error: command 'g' failed: No such file or directory ---------------------------------------------------------------------------- ----------------- From pearu at scipy.org Sun Jan 4 06:23:21 2004 From: pearu at scipy.org (Pearu Peterson) Date: Sun, 4 Jan 2004 05:23:21 -0600 (CST) Subject: [SciPy-user] weave moved to scipy_core in CVS Message-ID: Hi, weave module is now moved to scipy_core. This simplifies building scipy/chaco modules that need weave (that might not be installed when first checking out scipy/chaco modules). So, when updating scipy or chaco_all or any other CVS module that contained weave then you'll probably see the following message: cvs server: cannot open directory /home/cvsroot/world/scipy/Lib/weave: \ No such file or directory In that case you should remove your local copy of weave (or if you have made changes to local weave then make a backup) and re-run 'cvs update' or 'cvs co weave'. Regards, Pearu From pajer at iname.com Sun Jan 4 09:24:48 2004 From: pajer at iname.com (Gary Pajer) Date: Sun, 4 Jan 2004 09:24:48 -0500 Subject: [SciPy-user] weave moved to scipy_core in CVS References: Message-ID: <001a01c3d2ce$82657910$01fd5644@playroom> > Hi, > > weave module is now moved to scipy_core. This simplifies building > scipy/chaco modules that need weave (that might not be installed > when first checking out scipy/chaco modules). Pearu, setup_no_chaco.py has vanished. -gary From pearu at cens.ioc.ee Sun Jan 4 10:18:53 2004 From: pearu at cens.ioc.ee (Pearu Peterson) Date: Sun, 4 Jan 2004 17:18:53 +0200 (EET) Subject: [SciPy-user] weave moved to scipy_core in CVS In-Reply-To: <001a01c3d2ce$82657910$01fd5644@playroom> Message-ID: On Sun, 4 Jan 2004, Gary Pajer wrote: > > Hi, > > > > weave module is now moved to scipy_core. This simplifies building > > scipy/chaco modules that need weave (that might not be installed > > when first checking out scipy/chaco modules). > > setup_no_chaco.py has vanished. Yes. Use setup.py instead. Chaco won't be build under scipy anymore. Pearu From bgoli at sun.ac.za Mon Jan 5 15:56:23 2004 From: bgoli at sun.ac.za (Brett G. Olivier) Date: Mon, 5 Jan 2004 22:56:23 +0200 (SAST) Subject: [SciPy-user] compiling on Windows: getting closer In-Reply-To: <002101c3d254$4b350e80$01fd5644@playroom> References: <002101c3d254$4b350e80$01fd5644@playroom> Message-ID: <1067.196.30.180.47.1073336183.squirrel@glue.jjj.sun.ac.za> Hi I have also been trying to compile scipy under windows using MinGW and experienced similar problems. However, inspired by this thread, I seem to have found a way that works with my set up. I've included build environment details at the end of this message. Here is the procedure I followed: ++++++++ Compiling scipy on windows ++++++++++++ First, I uncomment the ignore packages bit in setup.py to not build sparse and chaco and set up the atlas path. Then try to build with: python setup.py build --compiler=mingw32 This fails with the compiler error as discussed in this thread: "Lib\special\specfun\specfun.f:1513: internal compiler error: in emit_swap_ins at reg-stack.c:987" Some preliminary investigation has revealed that the -O2 and -O3 flag seems to cause the compiler error when building specfun.f so I go to Lib\special\specfun and build specfun.f without optimization (removing the offending functions allows a successful compile but leads to many scary link errors later on). I just used the setup compile call without the -O3 flag: g77 -Wall -fno-second-underscore -funroll-loops -march=pentium3 -malign-double -fomit-frame-pointer -c specfun.f -o specfun.o this built a specfun.o which I copied to: build\temp.win32-2.3 rebuild: python setup.py build --compiler=mingw32 specfun.f should now build but setup fails with: "g -shared build\temp.win32-2.3\Release\lib\cluster\src\vq_wrap.o -Lc:\python23\libs -Lc:\python23\PCBuild -Lbuild\temp.win32-2.3 -lpython23 -lrootfind -lc_misc -lcephes -o build\lib.win32-2.3\scipy\cluster\_vq.pyd error: command 'g' failed: No such file or directory" I am still trying to find where this call is made to patch, but in the meantime I went to the mingw\bin directory (in my case c:\mingw\bin) and copied g++.exe to g.exe rebuild: python setup.py build --compiler=mingw32 Scipy should now build to completion. install: python setup.py build --compiler=mingw32 install +++++++ I'm sure there is a better way of doing this but I managed to get a working installation, which (after running them a few times) passed all the scipy.test() tests. I'll try this with the latest cvs snapshot sometime soon. My evironment is win2000, Python 2.3, mingw 3.1 plus latest updates (uses gcc 3.3.1), Atlas 3.6.0 built with cygwin (gcc 3.3.1). A relatively recent f2py cvs version, Numeric 23.1 built with mingw and scipy cvs version '0.2.1_251.4498'. Hope this helps a bit Brett >> Could you try to find a work around to this compiler bug? >> Or just comment the offending function in specfun.f out in >> order to complete scipy completion. >> >> Pearu > > Compiling on Windows, continued. > I short-circuited the functions in specfun.f that were causing problems. > Those functions are RMN2L and ASWFB, the radial and angular prolate and > oblate spherical functions of the first kind. > Hopefully, I've not created a land mine. Now specfun.f compiles to > completion. > > But I have a new problem, one that looks odd to my eye. See below. Is > there a command called "g" ? > > BTW, if anyone can tell me that they've built scipy on Windows, and how > they > did it, please contact me. > How did the Enthought people do it? > > -gary -- Brett G. Olivier (bgoli at sun.ac.za) Biochemistry Dept, Stellenbosch University Tel: +27 21 8085871 Fax: +27 21 8085863 From pajer at iname.com Mon Jan 5 18:25:12 2004 From: pajer at iname.com (Gary Pajer) Date: Mon, 5 Jan 2004 18:25:12 -0500 Subject: [SciPy-user] compiling on Windows: getting closer References: <002101c3d254$4b350e80$01fd5644@playroom> <1067.196.30.180.47.1073336183.squirrel@glue.jjj.sun.ac.za> Message-ID: <001d01c3d3e3$2a9cc560$01fd5644@playroom> copying g++.exe to g.exe ... brilliant. Now it compiles to completion. To install I had to use python setup.py install --skip_build Scipy seems to work, mostly. I recompiled the original specfun.f as you suggested, but I'm not altogether sure that the new specfun.o was linked in. How can I ensure this? .. I guess I could see if the two functions I shorted are working... A few problems: scipy.test() gives a few errors, xplt doesn't seem to work, plt doesn't seem to work, gplt *does* work. I'll look into these errors and report back. ( for reference I'll report how I shorted out the functions.) > Hi > > I have also been trying to compile scipy under windows using MinGW and > experienced similar problems. However, inspired by this thread, I seem to > have found a way that works with my set up. I've included build > environment details at the end of this message. > > Here is the procedure I followed: > > ++++++++ Compiling scipy on windows ++++++++++++ > > First, I uncomment the ignore packages bit in setup.py to not build sparse > and chaco and set up the atlas path. Then try to build with: > > python setup.py build --compiler=mingw32 > > This fails with the compiler error as discussed in this thread: > > "Lib\special\specfun\specfun.f:1513: internal compiler error: in > emit_swap_ins > at reg-stack.c:987" > > Some preliminary investigation has revealed that the -O2 and -O3 flag > seems to cause the compiler error when building specfun.f so I go to > Lib\special\specfun and build specfun.f without optimization (removing the > offending functions allows a successful compile but leads to many scary > link errors later on). I just used the setup compile call without the -O3 > flag: > > g77 -Wall -fno-second-underscore -funroll-loops -march=pentium3 > -malign-double -fomit-frame-pointer -c specfun.f -o specfun.o > > this built a specfun.o which I copied to: build\temp.win32-2.3 > > rebuild: python setup.py build --compiler=mingw32 > > specfun.f should now build but setup fails with: > > "g -shared build\temp.win32-2.3\Release\lib\cluster\src\vq_wrap.o > -Lc:\python23\libs -Lc:\python23\PCBuild -Lbuild\temp.win32-2.3 -lpython23 > -lrootfind -lc_misc -lcephes -o build\lib.win32-2.3\scipy\cluster\_vq.pyd > error: command 'g' failed: No such file or directory" > > I am still trying to find where this call is made to patch, but in the > meantime I went to the mingw\bin directory (in my case c:\mingw\bin) and > > copied g++.exe to g.exe > > rebuild: python setup.py build --compiler=mingw32 > > Scipy should now build to completion. > > install: python setup.py build --compiler=mingw32 install > > +++++++ > > I'm sure there is a better way of doing this but I managed to get a > working installation, which (after running them a few times) passed all > the scipy.test() tests. I'll try this with the latest cvs snapshot > sometime soon. > > My evironment is win2000, Python 2.3, mingw 3.1 plus latest updates (uses > gcc 3.3.1), Atlas 3.6.0 built with cygwin (gcc 3.3.1). A relatively recent > f2py cvs version, Numeric 23.1 built with mingw and scipy cvs version > '0.2.1_251.4498'. > > Hope this helps a bit > Brett > > > >> Could you try to find a work around to this compiler bug? > >> Or just comment the offending function in specfun.f out in > >> order to complete scipy completion. > >> > >> Pearu > > > > Compiling on Windows, continued. > > I short-circuited the functions in specfun.f that were causing problems. > > Those functions are RMN2L and ASWFB, the radial and angular prolate and > > oblate spherical functions of the first kind. > > Hopefully, I've not created a land mine. Now specfun.f compiles to > > completion. > > > > But I have a new problem, one that looks odd to my eye. See below. Is > > there a command called "g" ? > > > > BTW, if anyone can tell me that they've built scipy on Windows, and how > > they > > did it, please contact me. > > How did the Enthought people do it? > > > > -gary > > -- > Brett G. Olivier (bgoli at sun.ac.za) > > Biochemistry Dept, Stellenbosch University > Tel: +27 21 8085871 Fax: +27 21 8085863 > > _______________________________________________ > SciPy-user mailing list > SciPy-user at scipy.net > http://www.scipy.net/mailman/listinfo/scipy-user From pajer at iname.com Mon Jan 5 20:06:52 2004 From: pajer at iname.com (Gary Pajer) Date: Mon, 5 Jan 2004 20:06:52 -0500 Subject: [SciPy-user] How I compiled scipy on Windows References: <002101c3d254$4b350e80$01fd5644@playroom><1067.196.30.180.47.1073336183.squirrel@glue.jjj.sun.ac.za> <001d01c3d3e3$2a9cc560$01fd5644@playroom> Message-ID: <000d01c3d3f1$5ece8e00$01fd5644@playroom> I had intended to figure out how to compile scipy on Windows, and then write a complete guide. That's not what happened. I have a guide, but it's not complete, and it doesn't produce a fully functional scipy (it's close, though). The road was not straight, and I've forgotten some of the steps. Some of steps may be out of order (not sure). But I publish it here because it might be helpful as is, and so that someone else can distill it and fix it up. If anyone has comments, questions, corrections, or solutions, please post them. -gary ----------------------------------------------------------- This is how I got scipy compiled. It is no doubt overly complicated, but it's what I did. In particular, I used two different compilers ... cygwin gcc for ATLAS and LAPACK, and MinGW for scipy. Could this be the source of lingering problems?? I started with the Enthough Edition of python 2.3. MinGW: I installed the latest version in case fixed the bug referred to below. It didn't, so it was probably unneccesary. I installed it in c:\mingw. (I installed everything into paths without spaces to avoid that problem) CYGWIN: Make sure you have the latest version of cygwin. An easy installation. Go to http://www.cywin.org and follow the installation instructions. Do yourself a favor and do a "complete" installation. Make sure you have gcc-devel. Make sure you have g77 (a package separate from gcc) FFTW: I installed this a while ago, and I don't remember how I did it. Since I don't remember, it must have gone smoothly. I installed it into c:\fftw. ATLAS: Get the *latest* sources from http://math-atlas.sourceforge.net Move everything to c:\atlas execute "make" answer all questions, choosing the default most of the time take note of the name of the architecture that atlas has come up with. in my case, it was WinNT_PIIISSE1 execute "make install arch=XXXXX" where XXXXX is the name of the archtecture that atlas built for. WinNT_PIIISSE1 in my case. Go get lunch. (I moved the libraries to c:\atlib for my own convenience. See the environment variables below under COMPILING SCIPY) Note: many people have had trouble compiling atlas under cygwin. If you have the *newest* cygwin (and compilers) and the *newest* atlas, you shouldn't have these problems. Note: you may see warnings about performance degredation when compiling with gcc 3.1 along with advice to use gcc 2.95. Forget that advice; it's outdated if you have the *latest* versions of everything. You are probably left with an *incomplete* lapack library. Check the size of the library (in my case it is c:\atlas\lib\WinNT_PIIISSE1\liblapack.a) If it's around 6 MB, you are somehow lucky. It's probably around 250 kB. You have to fix it. First, install LAPACK. LAPACK: Get lapack.tgz from I unpacked to c:\lapack Rename c:\lapack\make.inc to c:\lapack\make.inc.original (or something) In c:\lapack\INSTALL directory, there is a file make.inc.linux. Move and rename to c:\lapack\make.inc Execute "make lib" (the linux make.inc works for cygwin.) This creates c:\lapack\lapack_LINUX.a, which is a complete lapack library. Now we have to dig into this library and put in the atlas files that we compliled earlier. The instructions for how to do this are in the FAQ on the atlas website: http://math-atlas.sourceforge.net/errata.html#completelp It amounts to cd'ing to your lapack path, and mkdir tmp cd tmp ar x ../liblapack.a cp ../liblapack.a ar r ../liblapack.a *.o cd .. rm -rf tmp BUGS WORK-AROUND: Something is screwy in the setup script, and at one point the build tries to compile a file using the command "g". Got around this one by copying g++.exe to g.exe (thanks to Brett Oliver) Because of a compiler bug, two functions in specfun.f won't compile, and the install crashes. There are two possible fixes. I don't know which one is better. First fix: Brett Oliver had success by changing an optimization flag. Go to the directory containing specfun.f (.../scipy/Lib/special/specfun) and compile specfun.f by hand (without the -O3 flag): g77 -Wall -fno-second-underscore -funroll-loops -march=pentium3 -malign-double -fomit-frame-pointer -c specfun.f -o specfun.o this built a specfun.o which I copied to: build\temp.win32-2.3 goto COMPILING SCIPY Second possible fix: Short-circuit the two offending functions. I short-circuit them by having them return meaningless data. I don't just comment the functions out (that doesn't work). ***NB*** I am not at all sure that this doesn't break something else. It certainly breaks these functions, so it's obviously not ideal. 1. the function RMN2L, which starts at line 1439 in specfun.f I replace this function with the following: ---------------------------- SUBROUTINE RMN2L(M,N,C,X,DF,KD,R2F,R2D,ID) IMPLICIT DOUBLE PRECISION (A-H,O-Z) ID=1 R2D=1. R2F=1. RETURN END ---------------------------- 2. the function ASWFB, which starts at line 7837 in specfun.f I replace this function with the following: ----------------------------- SUBROUTINE ASWFB(M,N,C,X,KD,CV,S1F,S1D) IMPLICIT DOUBLE PRECISION (A-H,O-Z) S1F=0.0D0 S1D=0.0D0 RETURN END ----------------------------- COMPILING SCIPY: set environment variables set FFTW=c:\fftw set ATLAS=c:\atlib (I moved the ATLAS libraries to this directory) set SWIG=c:\swig (I don't know if this is necessary. I tried it at some point and left it in) use command python setup.py --compiler=mingw32 cross your fingers. Note: the first time I did this, python couldn't find something... unfortunately, I neglected to take notes. I think it was libpython23.a I'm sorry that I can't be more specific about this. In any event, I now have a file called c:\python23\libs\libpython23.a Check to see if you have one. INSTALLING: python setup.py install --skip-build In my case, scipy mostly works. scipy.test() returns a few errors (one at the 100% level). plt and xplt don't work. gplt does work. I'm still working on it. From pajer at iname.com Mon Jan 5 20:50:48 2004 From: pajer at iname.com (Gary Pajer) Date: Mon, 5 Jan 2004 20:50:48 -0500 Subject: [SciPy-user] How I compiled scipy on Windows (small correction) References: <002101c3d254$4b350e80$01fd5644@playroom><1067.196.30.180.47.1073336183.squirrel@glue.jjj.sun.ac.za><001d01c3d3e3$2a9cc560$01fd5644@playroom> <000d01c3d3f1$5ece8e00$01fd5644@playroom> Message-ID: <000501c3d3f7$82365ed0$01fd5644@playroom> Small correction already: > > I started with the Enthough Edition of python 2.3. > ... and updated with the latest binary installer from python.org From pajer at iname.com Mon Jan 5 22:21:13 2004 From: pajer at iname.com (Gary Pajer) Date: Mon, 5 Jan 2004 22:21:13 -0500 Subject: [SciPy-user] How I compiled scipy on Windows (small correction) References: <002101c3d254$4b350e80$01fd5644@playroom><1067.196.30.180.47.1073336183.squirrel@glue.jjj.sun.ac.za><001d01c3d3e3$2a9cc560$01fd5644@playroom><000d01c3d3f1$5ece8e00$01fd5644@playroom> <000501c3d3f7$82365ed0$01fd5644@playroom> Message-ID: <002901c3d404$238e4570$01fd5644@playroom> Error report: scipy.plt *does* work. I forgot gui_thread until now. scipy.test() returns a different number of errors each time I run it. Sometimes it reports zero errors. As many as six. Usually two. The error that occurs most often involves finding the zeros of Fresnel functions. I seem to have a reasonably well-functioning up-to-date scipy. From shuntim.luk at polyu.edu.hk Tue Jan 6 00:31:32 2004 From: shuntim.luk at polyu.edu.hk (LUK ShunTim) Date: Tue, 06 Jan 2004 13:31:32 +0800 Subject: [SciPy-user] How I compiled scipy on Windows (small correction) In-Reply-To: <000501c3d3f7$82365ed0$01fd5644@playroom> References: <002101c3d254$4b350e80$01fd5644@playroom><1067.196.30.180.47.1073336183.squirrel@glue.jjj.sun.ac.za><001d01c3d3e3$2a9cc560$01fd5644@playroom> <000d01c3d3f1$5ece8e00$01fd5644@playroom> <000501c3d3f7$82365ed0$01fd5644@playroom> Message-ID: <3FFA4834.1010808@polyu.edu.hk> Gary Pajer wrote: > Small correction already: > > > >>I started with the Enthough Edition of python 2.3. >> > > > ... and updated with the latest binary installer from python.org > Can you share your experience in this replacement/update? Does it break anything (so far)? Regards, ST -- From pajer at iname.com Tue Jan 6 09:06:29 2004 From: pajer at iname.com (Gary Pajer) Date: Tue, 6 Jan 2004 09:06:29 -0500 Subject: [SciPy-user] How I compiled scipy on Windows (small correction) References: <002101c3d254$4b350e80$01fd5644@playroom><1067.196.30.180.47.1073336183.squirrel@glue.jjj.sun.ac.za><001d01c3d3e3$2a9cc560$01fd5644@playroom><000d01c3d3f1$5ece8e00$01fd5644@playroom><000501c3d3f7$82365ed0$01fd5644@playroom> <3FFA4834.1010808@polyu.edu.hk> Message-ID: <000501c3d45e$4825ff90$01fd5644@playroom> > Can you share your experience in this replacement/update? Does it break > anything (so far)? After less than 24 hours, and minimal testing, no significant problems. I have run across an instance of the well-known "float deprecation warning" while using scipy.plt. This warning is unrelated to the Windows build. As time permits I'll track it down and fix it. I'll note that I am using Brett Oliver's procedure of compiling specfun.f separately, without the -O3 flag. -gary From rodrigo.vivi at ic.unicamp.br Thu Jan 8 08:25:00 2004 From: rodrigo.vivi at ic.unicamp.br (Rodrigo Vivi) Date: Thu, 8 Jan 2004 11:25:00 -0200 Subject: [SciPy-user] I need help ! URGENT Message-ID: <200401081125.00586.rodrigo.vivi@ic.unicamp.br> I'm a brazilian researcher working in a governmental research center. I'm trying to use SciPy for the first time. This is the result: [vivi at lajinha vivi]$ python Python 2.3 (#2, Aug 31 2003, 17:27:29) [GCC 3.3.1 (Mandrake Linux 9.2 3.3.1-1mdk)] on linux2 Type "help", "copyright", "credits" or "license" for more information. >>> from scipy import * /usr/lib/python2.3/site-packages/scipy_base/__init__.py:117: RuntimeWarning: Python C API version mismatch for module fastumath: This Python has API version 1012, module fastumath has version 1011. import fastumath # no need to use scipy_base.fastumath Segmentation fault [vivi at lajinha vivi]$ I Can't remove Python 2.3 and install Python 2.2. What can I do? __________________________________________ Rodrigo Vivi Computer Engeenier Cenpra - Centro de Pesquisas Renato Archer http://www.cenpra.gov.br __________________________________________ From arnd.baecker at web.de Thu Jan 8 08:30:58 2004 From: arnd.baecker at web.de (Arnd Baecker) Date: Thu, 8 Jan 2004 14:30:58 +0100 (CET) Subject: [SciPy-user] I need help ! URGENT In-Reply-To: <200401081125.00586.rodrigo.vivi@ic.unicamp.br> References: <200401081125.00586.rodrigo.vivi@ic.unicamp.br> Message-ID: Hi, somehow this looks to me as if scipy was complied for a different python version (but I might well be wrong ...). Can you elaborate more on how you installed python and scipy? Arnd On Thu, 8 Jan 2004, Rodrigo Vivi wrote: > I'm a brazilian researcher working in a governmental research center. > > I'm trying to use SciPy for the first time. > This is the result: > > [vivi at lajinha vivi]$ python > Python 2.3 (#2, Aug 31 2003, 17:27:29) > [GCC 3.3.1 (Mandrake Linux 9.2 3.3.1-1mdk)] on linux2 > Type "help", "copyright", "credits" or "license" for more information. > >>> from scipy import * > /usr/lib/python2.3/site-packages/scipy_base/__init__.py:117: RuntimeWarning: > Python C API version mismatch for module fastumath: This Python has API > version 1012, module fastumath has version 1011. > import fastumath # no need to use scipy_base.fastumath > Segmentation fault > [vivi at lajinha vivi]$ > > > I Can't remove Python 2.3 and install Python 2.2. > What can I do? > > > __________________________________________ > > Rodrigo Vivi > Computer Engeenier > Cenpra - Centro de Pesquisas Renato Archer > http://www.cenpra.gov.br > __________________________________________ > > _______________________________________________ > SciPy-user mailing list > SciPy-user at scipy.net > http://www.scipy.net/mailman/listinfo/scipy-user > From rodrigo.vivi at ic.unicamp.br Thu Jan 8 08:46:32 2004 From: rodrigo.vivi at ic.unicamp.br (Rodrigo Vivi) Date: Thu, 8 Jan 2004 11:46:32 -0200 Subject: [SciPy-user] I need help ! URGENT In-Reply-To: References: <200401081125.00586.rodrigo.vivi@ic.unicamp.br> Message-ID: <200401081146.32530.rodrigo.vivi@ic.unicamp.br> I installed the python 2.3 with Mandrake Linux 9.2. So I downloaded SciPy-0.2.0_alpha_200.4161.linux2_py.tar.gz from www.scipy.org and I executed: $ cd /usr/lib/python2.3/site-packages $ tar -zxvf /home/vivi/downloads/SciPy-0.2.0_alpha_200.4161.linux2_py.tar.gz What more Must I do? In somewhere I read that scipy need python 2.2 but I can't change python version. __________________________________________ Rodrigo Vivi Computer Engeenier Cenpra - Centro de Pesquisas Renato Archer http://www.cenpra.gov.br __________________________________________ Em Qui 08 Jan 2004 11:30, Arnd Baecker escreveu: > Hi, > > somehow this looks to me as if scipy was complied for a different > python version (but I might well be wrong ...). > Can you elaborate more on how you installed python > and scipy? > > Arnd > > On Thu, 8 Jan 2004, Rodrigo Vivi wrote: > > I'm a brazilian researcher working in a governmental research center. > > > > I'm trying to use SciPy for the first time. > > This is the result: > > > > [vivi at lajinha vivi]$ python > > Python 2.3 (#2, Aug 31 2003, 17:27:29) > > [GCC 3.3.1 (Mandrake Linux 9.2 3.3.1-1mdk)] on linux2 > > Type "help", "copyright", "credits" or "license" for more information. > > > > >>> from scipy import * > > > > /usr/lib/python2.3/site-packages/scipy_base/__init__.py:117: > > RuntimeWarning: Python C API version mismatch for module fastumath: This > > Python has API version 1012, module fastumath has version 1011. > > import fastumath # no need to use scipy_base.fastumath > > Segmentation fault > > [vivi at lajinha vivi]$ > > > > > > I Can't remove Python 2.3 and install Python 2.2. > > What can I do? > > > > > > __________________________________________ > > > > Rodrigo Vivi > > Computer Engeenier > > Cenpra - Centro de Pesquisas Renato Archer > > http://www.cenpra.gov.br > > __________________________________________ > > > > _______________________________________________ > > SciPy-user mailing list > > SciPy-user at scipy.net > > http://www.scipy.net/mailman/listinfo/scipy-user From pearu at scipy.org Thu Jan 8 08:59:37 2004 From: pearu at scipy.org (Pearu Peterson) Date: Thu, 8 Jan 2004 07:59:37 -0600 (CST) Subject: [SciPy-user] I need help ! URGENT In-Reply-To: <200401081146.32530.rodrigo.vivi@ic.unicamp.br> Message-ID: On Thu, 8 Jan 2004, Rodrigo Vivi wrote: > I installed the python 2.3 with Mandrake Linux 9.2. > So I downloaded SciPy-0.2.0_alpha_200.4161.linux2_py.tar.gz from www.scipy.org > and I executed: > $ cd /usr/lib/python2.3/site-packages > $ tar -zxvf /home/vivi/downloads/SciPy-0.2.0_alpha_200.4161.linux2_py.tar.gz > > What more Must I do? > > In somewhere I read that scipy need python 2.2 but I can't change python > version. I suggest getting scipy from CVS and building it yourself. Follow the instructions in INSTALL.txt. Regards, Pearu From arnd.baecker at web.de Thu Jan 8 09:02:13 2004 From: arnd.baecker at web.de (Arnd Baecker) Date: Thu, 8 Jan 2004 15:02:13 +0100 (CET) Subject: [SciPy-user] I need help ! URGENT In-Reply-To: <200401081146.32530.rodrigo.vivi@ic.unicamp.br> References: <200401081125.00586.rodrigo.vivi@ic.unicamp.br> <200401081146.32530.rodrigo.vivi@ic.unicamp.br> Message-ID: On Thu, 8 Jan 2004, Rodrigo Vivi wrote: > I installed the python 2.3 with Mandrake Linux 9.2. > So I downloaded SciPy-0.2.0_alpha_200.4161.linux2_py.tar.gz from www.scipy.org > and I executed: > $ cd /usr/lib/python2.3/site-packages > $ tar -zxvf /home/vivi/downloads/SciPy-0.2.0_alpha_200.4161.linux2_py.tar.gz Hmm, the build log indicates to me (but maybe I am wrong) that Python-2.1.3 was used for the build. > What more Must I do? Honestly, I would recommend to build scipy from source, at least that's what I always do. (I will send you my sketchy notes on this privately). Best, Arnd From nwagner at mecha.uni-stuttgart.de Sun Jan 11 13:49:24 2004 From: nwagner at mecha.uni-stuttgart.de (nwagner at mecha.uni-stuttgart.de) Date: Sun, 11 Jan 2004 19:49:24 +0100 (CET) Subject: [SciPy-user] Interessant - und bei springeronline.com zu finden. Message-ID: <7741848.1073846964422.JavaMail.avbeawls@desws021> From nwagner at mecha.uni-stuttgart.de Tue Jan 13 04:20:31 2004 From: nwagner at mecha.uni-stuttgart.de (Nils Wagner) Date: Tue, 13 Jan 2004 10:20:31 +0100 Subject: [SciPy-user] Subdivision of an interval Message-ID: <4003B85F.A7A043D3@mecha.uni-stuttgart.de> Hi all, How can I subdivide an interval provided that size(x) = n ? n = 10 h = (xu-xl)/n x = arange(xl,xu,h) Sometimes size(x) <> n Nils From pearu at scipy.org Tue Jan 13 05:07:42 2004 From: pearu at scipy.org (Pearu Peterson) Date: Tue, 13 Jan 2004 04:07:42 -0600 (CST) Subject: [SciPy-user] Subdivision of an interval In-Reply-To: <4003B85F.A7A043D3@mecha.uni-stuttgart.de> Message-ID: On Tue, 13 Jan 2004, Nils Wagner wrote: > Hi all, > > How can I subdivide an interval provided that size(x) = n ? > > n = 10 > h = (xu-xl)/n > x = arange(xl,xu,h) > > Sometimes size(x) <> n Use: x=xl+h*arange(n) Pearu From jb1 at alfred.edu Wed Jan 14 03:01:21 2004 From: jb1 at alfred.edu (John Byrnes) Date: Wed, 14 Jan 2004 03:01:21 -0500 Subject: [SciPy-user] Re: Pythonw crashes if I try to plot Complex data points In-Reply-To: References: Message-ID: >>>> from scipy import * >>>> a = arange(1,10,0.2) >>>> b = a + 2.j >>>> b All works OK. >>>> gplt.plot(b) Crashes for me too... However, you forgot to tell scipy to import gplt. In any case it doesn't help anyway. Try this from scipy import * from scipy import plt a = arange(1,10,0.2) b = a + 2.j plt.plot(b) This snippet works fine for me. John From nwagner at mecha.uni-stuttgart.de Thu Jan 15 06:20:13 2004 From: nwagner at mecha.uni-stuttgart.de (Nils Wagner) Date: Thu, 15 Jan 2004 12:20:13 +0100 Subject: [SciPy-user] Representation of floating-point numbers Message-ID: <4006776D.3F2329F1@mecha.uni-stuttgart.de> Hi all, Python 2.3+ (#1, Sep 23 2003, 23:07:16) [GCC 3.3.1 (SuSE Linux)] on linux2 Type "help", "copyright", "credits" or "license" for more information. >>> 5.1 5.0999999999999996 >>> What is the reason for this strange behavior ? GNU Octave, version 2.1.49 (i686-pc-linux-gnu). Copyright (C) 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003 John W. Eaton. This is free software; see the source code for copying conditions. There is ABSOLUTELY NO WARRANTY; not even for MERCHANTIBILITY or FITNESS FOR A PARTICULAR PURPOSE. For details, type `warranty'. Please contribute if you find this software useful. For more information, visit http://www.octave.org/help-wanted.html Report bugs to . octave:1> format long octave:2> 5.1 ans = 5.10000000000000 Please can someone expand on the different behavior of python and octave. Nils From arnd.baecker at web.de Thu Jan 15 06:31:13 2004 From: arnd.baecker at web.de (Arnd Baecker) Date: Thu, 15 Jan 2004 12:31:13 +0100 (CET) Subject: [SciPy-user] Representation of floating-point numbers In-Reply-To: <4006776D.3F2329F1@mecha.uni-stuttgart.de> References: <4006776D.3F2329F1@mecha.uni-stuttgart.de> Message-ID: Hi Nils, On Thu, 15 Jan 2004, Nils Wagner wrote: > Hi all, > > Python 2.3+ (#1, Sep 23 2003, 23:07:16) > [GCC 3.3.1 (SuSE Linux)] on linux2 > Type "help", "copyright", "credits" or "license" for more information. > >>> 5.1 > 5.0999999999999996 > >>> > > What is the reason for this strange behavior ? Have look at the Python Tutorial, http://www.python.org/doc/current/tut/node15.html > octave:1> format long > octave:2> 5.1 > ans = 5.10000000000000 > > Please can someone expand on the different behavior of python and > octave. Notice that the number of digits is different, see below for some examples There might be a way to change the representation in octave. Arnd P.S.: I hope the difference between 5.0999999999999996 and 5.1 does not matter for your numerics ;-) In [1]: print 5.1 5.1 In [2]: 5.1 Out[2]: 5.0999999999999996 In [3]: print "%10.9f", 5.1 %10.9f 5.1 In [4]: print "%10.9f" %( 5.1) 5.100000000 In [5]: print "%f" %( 5.1) 5.100000 In [6]: print "%g" %( 5.1) 5.1 In [7]: print "%16.15f" %( 5.1) 5.100000000000000 In [8]: print "%18.16f" %( 5.1) 5.0999999999999996 From nwagner at mecha.uni-stuttgart.de Fri Jan 16 02:49:01 2004 From: nwagner at mecha.uni-stuttgart.de (Nils Wagner) Date: Fri, 16 Jan 2004 08:49:01 +0100 Subject: [SciPy-user] LU decomposition of a sparse matrix ? Message-ID: <4007976D.192F6AC1@mecha.uni-stuttgart.de> Hi all, from scipy import * from scipy.sparse import * def sylvester(n): d = zeros((2, n), Float) d[0, :] = arange(n, 0, -1) d[1, :] = arange(1, n+1) print d A = spdiags(d, [-1, 1], n, n) return A n = 6 S = sylvester(n) print S Is it possible to convert a sparse matrix into a full matrix in scipy ? How about LU decomposition in case of sparse matrices ? Nils From hetal75 at hotmail.com Fri Jan 16 08:01:55 2004 From: hetal75 at hotmail.com (Hetal Patel) Date: Fri, 16 Jan 2004 13:01:55 +0000 Subject: [SciPy-user] Re: Pythonw crashes if I try to plot Complex data points Message-ID: I had the same problem too. It apears that Gplt does not support he plotting of complex data points. >From: John Byrnes >Reply-To: SciPy Users List >To: scipy-user at scipy.org >Subject: [SciPy-user] Re: Pythonw crashes if I try to plot Complex data >points >Date: Wed, 14 Jan 2004 03:01:21 -0500 > > >>>>>from scipy import * >>>>>a = arange(1,10,0.2) >>>>>b = a + 2.j >>>>>b >All works OK. > >>>>>gplt.plot(b) >Crashes for me too... However, you forgot to tell scipy to import gplt. In >any case it doesn't help anyway. > >Try this > >from scipy import * >from scipy import plt >a = arange(1,10,0.2) >b = a + 2.j >plt.plot(b) > >This snippet works fine for me. > >John > >_______________________________________________ >SciPy-user mailing list >SciPy-user at scipy.net >http://www.scipy.net/mailman/listinfo/scipy-user _________________________________________________________________ Express yourself with cool new emoticons http://www.msn.co.uk/specials/myemo From oliphant at ee.byu.edu Fri Jan 16 17:47:45 2004 From: oliphant at ee.byu.edu (Travis E. Oliphant) Date: Fri, 16 Jan 2004 15:47:45 -0700 Subject: [SciPy-user] LU decomposition of a sparse matrix ? In-Reply-To: <4007976D.192F6AC1@mecha.uni-stuttgart.de> References: <4007976D.192F6AC1@mecha.uni-stuttgart.de> Message-ID: <40086A11.2050503@ee.byu.edu> Nils Wagner wrote: > Hi all, > > from scipy import * > from scipy.sparse import * > def sylvester(n): > d = zeros((2, n), Float) > d[0, :] = arange(n, 0, -1) > d[1, :] = arange(1, n+1) > print d > A = spdiags(d, [-1, 1], n, n) > return A > n = 6 > S = sylvester(n) > print S > > Is it possible to convert a sparse matrix into a full matrix in scipy ? There is not a canned function yet I don't think (but check the class definition). > > How about LU decomposition in case of sparse matrices ? > Look at the example of solving a sparse linear system (it does an LU decomposition first, I believe). -Travis From nwagner at mecha.uni-stuttgart.de Wed Jan 21 04:44:54 2004 From: nwagner at mecha.uni-stuttgart.de (Nils Wagner) Date: Wed, 21 Jan 2004 10:44:54 +0100 Subject: [SciPy-user] linalg.eig on sparse matrices Message-ID: <400E4A16.1F084543@mecha.uni-stuttgart.de> Hi all, AFAIK, scipy cannot handle sparse matrices in linalg.eig. A first step might be the conversion from sparse to full format. I am wondering if there is any tool for this task available ? Moreover, It would be great, to have eigs (an interface to ARPACK ) in scipy.. Any hint would be appreciated. Nils From j_r_fonseca at yahoo.co.uk Wed Jan 21 09:14:57 2004 From: j_r_fonseca at yahoo.co.uk (=?iso-8859-1?Q?Jos=E9?= Fonseca) Date: Wed, 21 Jan 2004 14:14:57 +0000 (UTC) Subject: [SciPy-user] Re: linalg.eig on sparse matrices References: <400E4A16.1F084543@mecha.uni-stuttgart.de> Message-ID: Nils, On Wed, 21 Jan 2004 10:44:54 +0100, Nils Wagner wrote: > Hi all, > > AFAIK, scipy cannot handle sparse matrices in linalg.eig. > A first step might be the conversion from sparse to full format. > I am wondering if there is any tool for this task available ? If scipy had support for sparse matrix multiplication (scipy has some sparse matrix support, but on CVS only AFAIK) then it would be quite straightforward using the ARPACK bindings in http://jrfonseca.dyndns.org/work/phd/#arpack . Almost a literal translation of MATLAB's 'eigs' M-file into Numeric Python would do. > Moreover, It would be great, to have eigs (an interface to ARPACK ) in > scipy.. The most import bit is the availability of a generic sparse matrix and the implementation of the basic operations (such as matrix multiplication, and matrix factorization). ARPACK is doesn't care about the actual sparse matrix format - it only wants a matrix-vector multiplication callback (and in some cases the vector mutliplication by the matrix inverse). For all purposes the matrix may not even exists fully in memory at any time (I actually used this once to determine the eigenvalues of a _full_ covariance matrix of a random field with millions of elements since it was cheaper to compute the matrix elements every time than to store it fully in memory). > Any hint would be appreciated. You can see at http://mefriss1.swan.ac.uk/svn/python/Sparse.py a real usage example. It implements a _very_ basic sparse matrix class (with addition, subtraction, and matrix-vector multiplication), and then the linear system solver, and a eigenvalue solver. It uses UMFPACK for matrix factorization and another C module - spblastk - for matrix-vector multiplication which can be a bottleneck. And in case you ask about integrating these ARPACK bindings with scipy: I don't see much point until scipy provides sparse matrices support, and I already have too much on my plate to take such an endeavor (and not much immediate interest either). Perhaps you could take this mater in your hands instead... I hope this helps. Jose Fonseca From h.jansen at fel.tno.nl Thu Jan 22 05:33:37 2004 From: h.jansen at fel.tno.nl (H Jansen) Date: 22 Jan 2004 11:33:37 +0100 Subject: [SciPy-user] Re: linalg.eig on sparse matrices In-Reply-To: References: <400E4A16.1F084543@mecha.uni-stuttgart.de> Message-ID: <1074767617.13121.46.camel@pc50002760.fel.tno.nl> One may want to have a look at pysparse: http://people.web.psi.ch/geus/pyfemax/pysparse.html Could this be a candidate for integration with the scipy package? -Henk On Wed, 2004-01-21 at 15:14, Jos? Fonseca wrote: > Nils, > > On Wed, 21 Jan 2004 10:44:54 +0100, Nils Wagner wrote: > > Hi all, > > > > AFAIK, scipy cannot handle sparse matrices in linalg.eig. > > A first step might be the conversion from sparse to full format. > > I am wondering if there is any tool for this task available ? > > If scipy had support for sparse matrix multiplication (scipy has some > sparse matrix support, but on CVS only AFAIK) then it would be quite > straightforward using the ARPACK bindings in > http://jrfonseca.dyndns.org/work/phd/#arpack . Almost a literal > translation of MATLAB's 'eigs' M-file into Numeric Python would do. > > > Moreover, It would be great, to have eigs (an interface to ARPACK ) in > > scipy.. > > The most import bit is the availability of a generic sparse matrix and > the implementation of the basic operations (such as matrix > multiplication, and matrix factorization). > > ARPACK is doesn't care about the actual sparse matrix format - it only > wants a matrix-vector multiplication callback (and in some cases the > vector mutliplication by the matrix inverse). For all purposes the > matrix may not even exists fully in memory at any time (I actually used > this once to determine the eigenvalues of a _full_ covariance matrix of > a random field with millions of elements since it was cheaper to compute > the matrix elements every time than to store it fully in memory). > > > Any hint would be appreciated. > > You can see at http://mefriss1.swan.ac.uk/svn/python/Sparse.py a real > usage example. It implements a _very_ basic sparse matrix class (with > addition, subtraction, and matrix-vector multiplication), and then the > linear system solver, and a eigenvalue solver. It uses UMFPACK for > matrix factorization and another C module - spblastk - for matrix-vector > multiplication which can be a bottleneck. > > And in case you ask about integrating these ARPACK bindings with scipy: > I don't see much point until scipy provides sparse matrices support, and > I already have too much on my plate to take such an endeavor (and not > much immediate interest either). Perhaps you could take this mater in > your hands instead... > > I hope this helps. > > Jose Fonseca > > _______________________________________________ > SciPy-user mailing list > SciPy-user at scipy.net > http://www.scipy.net/mailman/listinfo/scipy-user -- Henk Jansen TNO Physics and Electronics Laboratory -- ------------------------------------------------------------------------------ The disclaimer that applies to e-mail from TNO Physics and Electronics Laboratory can be found on: http://www.tno.nl/disclaimer/email.html ------------------------------------------------------------------------------ From oliphant at ee.byu.edu Thu Jan 22 14:54:07 2004 From: oliphant at ee.byu.edu (Travis E. Oliphant) Date: Thu, 22 Jan 2004 12:54:07 -0700 Subject: [SciPy-user] Re: linalg.eig on sparse matrices In-Reply-To: <1074767617.13121.46.camel@pc50002760.fel.tno.nl> References: <400E4A16.1F084543@mecha.uni-stuttgart.de> <1074767617.13121.46.camel@pc50002760.fel.tno.nl> Message-ID: <40102A5F.1040600@ee.byu.edu> H Jansen wrote: > One may want to have a look at pysparse: > > http://people.web.psi.ch/geus/pyfemax/pysparse.html Sparse matrices are definitely a candidate for SciPy. In fact, there is rudimentary support for Sparse matrices already there. The plan is to rewrite the Sparse matrix class into one (or more) Python type objects and support at least two different types of Sparse matrices: linked-list (for fast matrix formation), compressed sparse row (and compressed sparse column) for fast matrix manipulation. Four different data types are planned for these matrices. There is a project out there that gets a bit of the way there (I think it may be this one you speak of) but there is more to do to make it general purpose enough. If there are individuals out there with experience with Sparse matrices, this would be a great place to help out with development. -Travis Oliphant From h.jansen at fel.tno.nl Fri Jan 23 05:41:41 2004 From: h.jansen at fel.tno.nl (H Jansen) Date: 23 Jan 2004 11:41:41 +0100 Subject: [SciPy-user] Re: linalg.eig on sparse matrices In-Reply-To: <40102A5F.1040600@ee.byu.edu> References: <400E4A16.1F084543@mecha.uni-stuttgart.de> <1074767617.13121.46.camel@pc50002760.fel.tno.nl> <40102A5F.1040600@ee.byu.edu> Message-ID: <1074854501.13121.187.camel@pc50002760.fel.tno.nl> On Thu, 2004-01-22 at 20:54, Travis E. Oliphant wrote: > H Jansen wrote: > > One may want to have a look at pysparse: > > > > http://people.web.psi.ch/geus/pyfemax/pysparse.html > > Sparse matrices are definitely a candidate for SciPy. In fact, there is > rudimentary support for Sparse matrices already there. > > The plan is to rewrite the Sparse matrix class into one (or more) Python > type objects and support at least two different types of Sparse > matrices: linked-list (for fast matrix formation), compressed sparse > row (and compressed sparse column) for fast matrix manipulation. Four > different data types are planned for these matrices. > > There is a project out there that gets a bit of the way there (I think > it may be this one you speak of) but there is more to do to make it > general purpose enough. > > If there are individuals out there with experience with Sparse matrices, > this would be a great place to help out with development. I can't consider myself a "sparse matrix specialist" but I'm eager to test results and contribute to feature discussions. For instance, in pysparse index information is not available so you can't iterate over the sparse matrix elements --- this is a real handicap. Long time past (in my Fortran days) I've used Yousef Saad's Sparsekit which, if I remember correctly, was a pretty well featured sparse matrix kit. It didn't come integrated with iterative solvers of course since object-orientiation wasn't given much attention in the numerical computing community at that time (late 80's, early 90's). Things are now different, object-orientation has entered the numerical computation field (see e.g. http://oonumerics.org, http://www.numpy.org and many others ...). An interesting site (with article) is http://dynopt.cheme.cmu.edu/roscoe/RTOp/ from which I gather that the major challenge seems not to be a generic interface for dense and sparse storage (and associated linear direct and iterative solvers) but (1) the transparent (generic) and efficient access of in-core/out-of-core vector/array data; (2) the transparent (generic) and efficient access of distributively stored vector/array data; and (3) to facilitate efficient computation of (a growing number of) application specific matrix/vector operations without the need to extend the library with new methods. In Python the "arrayfrombuffer" module may provide the functionality needed for address (1). (... skipping issue (2) ...) In the article the proposed solution for (3) is to have each numerical library implement the "apply_op(...)" method which accepts user-defined (application specific) matrix/vector operators. I believe the same solution can work for Python: special-purpose vector/matrix operations can be implemented in a compilable language (Fortran/C++) and easily be interfaced with Python (using, for example, for Fortran: F2PY or pyfortran, and for C++: C++boost template library, PyCXX or Swig). These were just some ideas, -Henk > > -Travis Oliphant > > _______________________________________________ > SciPy-user mailing list > SciPy-user at scipy.net > http://www.scipy.net/mailman/listinfo/scipy-user -- Henk Jansen TNO Physics and Electronics Laboratory -- ------------------------------------------------------------------------------ The disclaimer that applies to e-mail from TNO Physics and Electronics Laboratory can be found on: http://www.tno.nl/disclaimer/email.html ------------------------------------------------------------------------------ From oliphant at ee.byu.edu Fri Jan 23 06:59:12 2004 From: oliphant at ee.byu.edu (Travis E. Oliphant) Date: Fri, 23 Jan 2004 04:59:12 -0700 Subject: [SciPy-user] Re: linalg.eig on sparse matrices In-Reply-To: <1074854501.13121.187.camel@pc50002760.fel.tno.nl> References: <400E4A16.1F084543@mecha.uni-stuttgart.de> <1074767617.13121.46.camel@pc50002760.fel.tno.nl> <40102A5F.1040600@ee.byu.edu> <1074854501.13121.187.camel@pc50002760.fel.tno.nl> Message-ID: <40110C90.7040208@ee.byu.edu> H Jansen wrote: > On Thu, 2004-01-22 at 20:54, Travis E. Oliphant wrote: > >>H Jansen wrote: >> >>>One may want to have a look at pysparse: >>> >>>http://people.web.psi.ch/geus/pyfemax/pysparse.html >> >>Sparse matrices are definitely a candidate for SciPy. In fact, there is >>rudimentary support for Sparse matrices already there. >> >>The plan is to rewrite the Sparse matrix class into one (or more) Python >>type objects and support at least two different types of Sparse >>matrices: linked-list (for fast matrix formation), compressed sparse >>row (and compressed sparse column) for fast matrix manipulation. Four >>different data types are planned for these matrices. >> >>There is a project out there that gets a bit of the way there (I think >>it may be this one you speak of) but there is more to do to make it >>general purpose enough. >> >>If there are individuals out there with experience with Sparse matrices, >>this would be a great place to help out with development. > > > I can't consider myself a "sparse matrix specialist" but I'm eager to > test results and contribute to feature discussions. For instance, in > pysparse index information is not available so you can't iterate over > the sparse matrix elements --- this is a real handicap. Long time past > (in my Fortran days) I've used Yousef Saad's Sparsekit which, if I > remember correctly, was a pretty well featured sparse matrix kit. It > didn't come integrated with iterative solvers of course since > object-orientiation wasn't given much attention in the numerical > computing community at that time (late 80's, early 90's). > Right now, the sparse matrix in SciPy uses Saad's toolkit (which is GPL and thus problematic for continuance in the SciPy tree). > In Python the "arrayfrombuffer" module may provide the functionality > needed for address (1). (... skipping issue (2) ...) In the article the > proposed solution for (3) is to have each numerical library implement > the "apply_op(...)" method which accepts user-defined (application > specific) matrix/vector operators. I believe the same solution can work > for Python: special-purpose vector/matrix operations can be implemented > in a compilable language (Fortran/C++) and easily be interfaced with > Python (using, for example, for Fortran: F2PY or pyfortran, and for C++: > C++boost template library, PyCXX or Swig). I think Python is a great tool for answering some of these questions, but right now that is not on my path. I would just like to see the sparse matrix object in SciPy improved. -Travis O. From tom at kornack.com Fri Jan 23 16:08:38 2004 From: tom at kornack.com (Tom Kornack) Date: Fri, 23 Jan 2004 16:08:38 -0500 Subject: [SciPy-user] Mac OS X 10.3 Installation? Message-ID: <509AB54A-4DE8-11D8-80E1-000A95DA14CE@kornack.com> Hello: Has anyone found a recipe for success installing scipy on 10.3? I notice that the scipy fink package hasn't yet been updated for 10.3. My various attempts have ended in failure. Here's the furthest I've gotten using the latest CVS of scipy: sudo python setup.py install ... ... building 'scipy.fftpack.convolve' extension gcc -L/sw/lib -bundle -flat_namespace -undefined suppress build/temp.darwin-7.2.0-PowerMacintosh-2.3/Lib/fftpack/src/convolve.o build/temp.darwin-7.2.0-PowerMacintosh-2.3/sw/lib/python2.3/site- packages/f2py2e/src/fortranobject.o build/temp.darwin-7.2.0-PowerMacintosh-2.3/build/temp.darwin-7.2.0- PowerMacintosh-2.3/convolvemodule.o -L/sw/lib -Lbuild/temp.darwin-7.2.0-PowerMacintosh-2.3 -Lbuild/temp.darwin-7.2.0-PowerMacintosh-2.3 -L/sw/lib/gcc/powerpc-apple-darwin7.2.0/3.4 -L/sw/lib -ldfftpack -ldrfftw -ldfftw -lrootfind -lsuperlu -lmyblas -lc_misc -lcephes -lg2c -o build/lib.darwin-7.2.0-PowerMacintosh-2.3/scipy/fftpack/convolve.so ld: build/temp.darwin-7.2.0-PowerMacintosh-2.3/build/temp.darwin-7.2.0- PowerMacintosh-2.3/convolvemodule.o has external relocation entries in non-writable section (__TEXT,__text) for symbols: restFP saveFP error: command 'gcc' failed with exit status 1 Any ideas what's happening here? I'm running as sudo. I'm not sure what is meant by a 'non-writable section'. I've tried the 0.2.0alpha release as well but don't get as far. Many thanks in advance for any guidance. Awfully sorry to be dumping my build problems on the list. Tom http://kornack.com Fundamental Symmetries Lab, Princeton University Thomas Kornack, 157 North Post Road, Princeton, NJ 08550-5009 609-716-7259 (h), 609-933-2186 (m), 609-258-0702 (w), 609-258-1625 (f) From chris at fonnesbeck.org Fri Jan 23 16:23:43 2004 From: chris at fonnesbeck.org (Christopher Fonnesbeck) Date: Fri, 23 Jan 2004 16:23:43 -0500 Subject: [SciPy-user] Mac OS X 10.3 Installation? In-Reply-To: <509AB54A-4DE8-11D8-80E1-000A95DA14CE@kornack.com> References: <509AB54A-4DE8-11D8-80E1-000A95DA14CE@kornack.com> Message-ID: <6BB08894-4DEA-11D8-8114-000A956FDAC0@fonnesbeck.org> On Jan 23, 2004, at 4:08 PM, Tom Kornack wrote: > Hello: > > Has anyone found a recipe for success installing scipy on 10.3? I > notice that the scipy fink package hasn't yet been updated for 10.3. > > My various attempts have ended in failure. Here's the furthest I've > gotten using the latest CVS of scipy: > > sudo python setup.py install > ... > ... > building 'scipy.fftpack.convolve' extension > gcc -L/sw/lib -bundle -flat_namespace -undefined suppress > build/temp.darwin-7.2.0-PowerMacintosh-2.3/Lib/fftpack/src/convolve.o > build/temp.darwin-7.2.0-PowerMacintosh-2.3/sw/lib/python2.3/site- > packages/f2py2e/src/fortranobject.o > build/temp.darwin-7.2.0-PowerMacintosh-2.3/build/temp.darwin-7.2.0- > PowerMacintosh-2.3/convolvemodule.o -L/sw/lib > -Lbuild/temp.darwin-7.2.0-PowerMacintosh-2.3 > -Lbuild/temp.darwin-7.2.0-PowerMacintosh-2.3 > -L/sw/lib/gcc/powerpc-apple-darwin7.2.0/3.4 -L/sw/lib -ldfftpack > -ldrfftw -ldfftw -lrootfind -lsuperlu -lmyblas -lc_misc -lcephes -lg2c > -o build/lib.darwin-7.2.0-PowerMacintosh-2.3/scipy/fftpack/convolve.so > ld: > build/temp.darwin-7.2.0-PowerMacintosh-2.3/build/temp.darwin-7.2.0- > PowerMacintosh-2.3/convolvemodule.o has external relocation entries in > non-writable section (__TEXT,__text) for symbols: > restFP > saveFP > error: command 'gcc' failed with exit status 1 > > Any ideas what's happening here? I'm running as sudo. I'm not sure > what is meant by a 'non-writable section'. I've tried the 0.2.0alpha > release as well but don't get as far. SciPy works great for me on OSX 10.3. A couple of hints: 1. Get /sw out of your library path. Nothing good comes of fink software when trying to compile SciPy. 2. Make sure you have the apple development tools installed. 3. Make sure the "ignore_packages" variable at the end of the setup.py file is uncommented. Hope this helps, Chris -- Christopher J. Fonnesbeck ( c h r i s @ f o n n e s b e c k . o r g ) Georgia Cooperative Fish & Wildlife Research Unit, University of Georgia From GHGweon at lbl.gov Fri Jan 23 21:08:14 2004 From: GHGweon at lbl.gov (Gey-Hong Gweon) Date: Fri, 23 Jan 2004 18:08:14 -0800 Subject: [SciPy-user] windows installation problem Message-ID: <4011D38E.5070502@lbl.gov> Hi, I am having a problem right at the beginning. I have downloaded and installed the binary package for windows - SciPy-0.2.1_253.3687.win32-py2.2. The following problem occurs. Could any one point me to a solution? Python 2.2.3 (#42, May 30 2003, 18:12:08) [MSC 32 bit (Intel)] on win32 Type "help", "copyright", "credits" or "license" for more information. >>> import scipy Traceback (most recent call last): File "", line 1, in ? File "C:\Python22\Lib\site-packages\scipy\__init__.py", line 11, in ? from scipy_base import * ImportError: No module named scipy_base Thanks, -- Gey-Hong Gweon, Ph.D. 510-486-6194 (Office: Room 2-202), 510-486-5530 (Fax) MS 2-200, Lawrence Berkely Lab, 1 Cyclotron Rd, Berkeley, CA 94720 From rkern at ucsd.edu Fri Jan 23 21:46:39 2004 From: rkern at ucsd.edu (Robert Kern) Date: Fri, 23 Jan 2004 18:46:39 -0800 Subject: [SciPy-user] Mac OS X 10.3 Installation? In-Reply-To: <6BB08894-4DEA-11D8-8114-000A956FDAC0@fonnesbeck.org> References: <509AB54A-4DE8-11D8-80E1-000A95DA14CE@kornack.com> <6BB08894-4DEA-11D8-8114-000A956FDAC0@fonnesbeck.org> Message-ID: <20040124024639.GA18274@taliesen.ucsd.edu> On Fri, Jan 23, 2004 at 04:23:43PM -0500, Christopher Fonnesbeck wrote: > On Jan 23, 2004, at 4:08 PM, Tom Kornack wrote: > > >Hello: > > > >Has anyone found a recipe for success installing scipy on 10.3? I > >notice that the scipy fink package hasn't yet been updated for 10.3. > > > >My various attempts have ended in failure. Here's the furthest I've > >gotten using the latest CVS of scipy: > > > >sudo python setup.py install > >... > >... > >building 'scipy.fftpack.convolve' extension > >gcc -L/sw/lib -bundle -flat_namespace -undefined suppress > >build/temp.darwin-7.2.0-PowerMacintosh-2.3/Lib/fftpack/src/convolve.o > >build/temp.darwin-7.2.0-PowerMacintosh-2.3/sw/lib/python2.3/site- > >packages/f2py2e/src/fortranobject.o > >build/temp.darwin-7.2.0-PowerMacintosh-2.3/build/temp.darwin-7.2.0- > >PowerMacintosh-2.3/convolvemodule.o -L/sw/lib > >-Lbuild/temp.darwin-7.2.0-PowerMacintosh-2.3 > >-Lbuild/temp.darwin-7.2.0-PowerMacintosh-2.3 > >-L/sw/lib/gcc/powerpc-apple-darwin7.2.0/3.4 -L/sw/lib -ldfftpack > >-ldrfftw -ldfftw -lrootfind -lsuperlu -lmyblas -lc_misc -lcephes -lg2c > >-o build/lib.darwin-7.2.0-PowerMacintosh-2.3/scipy/fftpack/convolve.so > >ld: > >build/temp.darwin-7.2.0-PowerMacintosh-2.3/build/temp.darwin-7.2.0- > >PowerMacintosh-2.3/convolvemodule.o has external relocation entries in > >non-writable section (__TEXT,__text) for symbols: > >restFP > >saveFP > >error: command 'gcc' failed with exit status 1 > > > >Any ideas what's happening here? I'm running as sudo. I'm not sure > >what is meant by a 'non-writable section'. I've tried the 0.2.0alpha > >release as well but don't get as far. Googling for "restFP saveFP" gave me the fix: add -lcc_dynamic to the link flags. > SciPy works great for me on OSX 10.3. A couple of hints: > > 1. Get /sw out of your library path. Nothing good comes of fink > software when trying to compile SciPy. Works fine for me using fink's python. This weekend, I'll try with MacPython. > 2. Make sure you have the apple development tools installed. > 3. Make sure the "ignore_packages" variable at the end of the setup.py > file is uncommented. I have a patch (hack) for the sparse package to make it compile. The kiva/chaco stuff seems to compile though I haven't tested running it. > Hope this helps, > Chris -- Robert Kern rkern at ucsd.edu "In the fields of hell where the grass grows high Are the graves of dreams allowed to die." -- Richard Harter From jbgao at turing.une.edu.au Sun Jan 4 18:32:49 2004 From: jbgao at turing.une.edu.au (Dr. Junbin Gao) Date: Sun, 04 Jan 2004 23:32:49 -0000 Subject: [SciPy-user] SciPy Installation Message-ID: <3FF8A202.7060306@turing.une.edu.au> Hi There I installed PyThon 2.2.3 on my D driver and downloaded sciPy SciPy-0.2.0_alpha_210.4077.win32-py2.2.exe But when I run the above exe file, I met the error message "Python version 2.2 required , which was not found in the registry" What does this mean?? Junbin Gao -------------- next part -------------- An HTML attachment was scrubbed... URL: From jbgao at turing.une.edu.au Sun Jan 4 18:57:20 2004 From: jbgao at turing.une.edu.au (Dr. Junbin Gao) Date: Sun, 04 Jan 2004 23:57:20 -0000 Subject: [SciPy-user] Further to my last email Message-ID: <3FF8A7D6.5050600@turing.une.edu.au> The version of PyThon I installed is 2.3. So now any SciPy version compatible to this one exists? Junbin From dr.nwagner at web.de Sat Jan 17 01:20:01 2004 From: dr.nwagner at web.de (Nils Wagner) Date: Sat, 17 Jan 2004 07:20:01 +0100 Subject: [SciPy-user] LU decomposition of a sparse matrix ? Message-ID: <200401170620.i0H6K0Q18056@mailgate5.cinetic.de> SciPy Users List schrieb am 16.01.04 23:47:06: Nils Wagner wrote: > Hi all, > > from scipy import * > from scipy.sparse import * > def sylvester(n): > d = zeros((2, n), Float) > d[0, :] = arange(n, 0, -1) > d[1, :] = arange(1, n+1) > print d > A = spdiags(d, [-1, 1], n, n) > return A > n = 6 > S = sylvester(n) > print S > > Is it possible to convert a sparse matrix into a full matrix in scipy ? There is not a canned function yet I don't think (but check the class definition). > > How about LU decomposition in case of sparse matrices ? > Look at the example of solving a sparse linear system (it does an LU decomposition first, I believe). def sparse_linear_solve(A,b): assert isspmatrix(A) assert A.storage=='CSR' gssv = eval('_superlu.' + A.ftype + 'gssv') return gssv(A.shape[0],A.shape[1],A.laste+1,A.data,A.index[0]-1,A.index[1]-1,b) splinsolve = sparse_linear_solve What is the meaning of gssv ? Nils -Travis _______________________________________________ SciPy-user mailing list SciPy-user at scipy.net http://www.scipy.net/mailman/listinfo/scipy-user ______________________________________________________________________________ Nachrichten, Musik und Spiele schnell und einfach per Quickstart im WEB.DE Screensaver - Gratis downloaden: http://screensaver.web.de/?mc=021110 From pajer at iname.com Sat Jan 24 13:51:02 2004 From: pajer at iname.com (Gary Pajer) Date: Sat, 24 Jan 2004 13:51:02 -0500 Subject: [SciPy-user] Further to my last email References: <3FF8A7D6.5050600@turing.une.edu.au> Message-ID: <004301c3e2ab$041b5150$01fd5644@playroom> > The version of PyThon I installed is 2.3. So now any SciPy version > compatible to this one exists? > > Junbin The simplest way is to install the "Enthought Edition" of python 2.3 which includes python 2.3 and a working scipy along with some other stuff. Download from http://www.enthought.com/ I don't know if it's necessary to uninstall your current python 2.3, but I think I would to be cautious. It is possible to compile scipy from source on Windows, but it's not a trivial exercise. I wouldn't recommend doing it until you have some reason to need it. I can send you some notes privately if you are interested. -g From tom at kornack.com Sat Jan 24 14:23:12 2004 From: tom at kornack.com (Tom Kornack) Date: Sat, 24 Jan 2004 14:23:12 -0500 Subject: [SciPy-user] Mac OS X 10.3 Installation? In-Reply-To: <20040124024639.GA18274@taliesen.ucsd.edu> References: <509AB54A-4DE8-11D8-80E1-000A95DA14CE@kornack.com> <6BB08894-4DEA-11D8-8114-000A956FDAC0@fonnesbeck.org> <20040124024639.GA18274@taliesen.ucsd.edu> Message-ID: Thanks to everyone who gave me hints. Here's what I needed to do: sudo setup.py build build_ext -lcc_dynamic install That's how to specify the link flag needed to get past the restFP saveFP issues. I also needed to uncomment the line to ignore 'sparse'. I hope this helps someone out there. Tom http://kornack.com Fundamental Symmetries Lab, Princeton University Thomas Kornack, 157 North Post Road, Princeton, NJ 08550-5009 609-716-7259 (h), 609-933-2186 (m), 609-258-0702 (w), 609-258-1625 (f) From russomf at hotmail.com Sun Jan 25 00:42:10 2004 From: russomf at hotmail.com (Mark Russo) Date: Sun, 25 Jan 2004 00:42:10 -0500 Subject: [SciPy-user] Error on import Message-ID: I have been using Python and wxPython very successfully for some time now. I installed SciPy, but get an error upon importing scipy (see following). I am using Windows 2000, Python 2.2.3 and wxPython 2.4.2.4. Thanks for your help. Mark _________________________ Microsoft Windows 2000 [Version 5.00.2195] (C) Copyright 1985-2000 Microsoft Corp. C:\>python ActivePython 2.2.3 Build 227 (ActiveState Corp.) based on Python 2.2.3 (#42, Nov 13 2003, 09:57:55) [MSC 32 bit (Intel)] on win32 Type "help", "copyright", "credits" or "license" for more information. >>>from scipy import * Traceback (most recent call last): File "", line 1, in ? File "C:\Python22\Lib\site-packages\scipy\__init__.py", line 11, in ? from scipy_base import * ImportError: No module named scipy_base >>> _________________________________________________________________ Let the new MSN Premium Internet Software make the most of your high-speed experience. http://join.msn.com/?pgmarket=en-us&page=byoa/prem&ST=1 From nwagner at mecha.uni-stuttgart.de Mon Jan 26 07:49:25 2004 From: nwagner at mecha.uni-stuttgart.de (Nils Wagner) Date: Mon, 26 Jan 2004 13:49:25 +0100 Subject: [SciPy-user] arbitrary precision arithmetic in scipy Message-ID: <40150CD5.196882CA@mecha.uni-stuttgart.de> Dear experts, Is it possible to use arbitrary precision arithmetic in scipy ? I am grateful for any hint. Nils From pearu at scipy.org Mon Jan 26 08:30:12 2004 From: pearu at scipy.org (Pearu Peterson) Date: Mon, 26 Jan 2004 07:30:12 -0600 (CST) Subject: [SciPy-user] arbitrary precision arithmetic in scipy In-Reply-To: <40150CD5.196882CA@mecha.uni-stuttgart.de> Message-ID: On Mon, 26 Jan 2004, Nils Wagner wrote: > Dear experts, > > Is it possible to use arbitrary precision arithmetic in scipy ? No (unless you're use integers) > I am grateful for any hint. http://gmpy.sourceforge.net/ Pearu From oliphant at ee.byu.edu Mon Jan 26 05:11:46 2004 From: oliphant at ee.byu.edu (Travis E. Oliphant) Date: Mon, 26 Jan 2004 03:11:46 -0700 Subject: [SciPy-user] windows installation problem In-Reply-To: <4011D38E.5070502@lbl.gov> References: <4011D38E.5070502@lbl.gov> Message-ID: <4014E7E2.2030205@ee.byu.edu> Gey-Hong Gweon wrote: > Hi, > > I am having a problem right at the beginning. I have downloaded and > installed the binary package for windows - > SciPy-0.2.1_253.3687.win32-py2.2. The following problem occurs. Could > any one point me to a solution? > > Python 2.2.3 (#42, May 30 2003, 18:12:08) [MSC 32 bit (Intel)] on win32 > Type "help", "copyright", "credits" or "license" for more information. > >>> import scipy > Traceback (most recent call last): > File "", line 1, in ? > File "C:\Python22\Lib\site-packages\scipy\__init__.py", line 11, in ? > from scipy_base import * > ImportError: No module named scipy_base > There are two binary packages created one is called SciPy-Core (it must be installed as well). Whoever put up the latest binary may not be aware of that (it currently builds under the scipy_core/build directory). -Travis O. From oliphant at ee.byu.edu Mon Jan 26 05:12:28 2004 From: oliphant at ee.byu.edu (Travis E. Oliphant) Date: Mon, 26 Jan 2004 03:12:28 -0700 Subject: [SciPy-user] windows installation problem In-Reply-To: <4011D38E.5070502@lbl.gov> References: <4011D38E.5070502@lbl.gov> Message-ID: <4014E80C.4060705@ee.byu.edu> Gey-Hong Gweon wrote: > Hi, > > I am having a problem right at the beginning. I have downloaded and > installed the binary package for windows - > SciPy-0.2.1_253.3687.win32-py2.2. The following problem occurs. Could > any one point me to a solution? > > Python 2.2.3 (#42, May 30 2003, 18:12:08) [MSC 32 bit (Intel)] on win32 > Type "help", "copyright", "credits" or "license" for more information. > >>> import scipy > Traceback (most recent call last): > File "", line 1, in ? > File "C:\Python22\Lib\site-packages\scipy\__init__.py", line 11, in ? > from scipy_base import * > ImportError: No module named scipy_base > > Thanks, I have a binary copy of the SciPy-core binary that I will try to make available on my SciPy page. -Travis O. From ehaux at ucmerced.edu Mon Jan 26 20:28:12 2004 From: ehaux at ucmerced.edu (ehaux at ucmerced.edu) Date: Mon, 26 Jan 2004 17:28:12 -0800 Subject: [SciPy-user] need help with gcc error Message-ID: <601d060407.60407601d0@ucmerced.edu> can anyone help with this?? i downloaded from cvs and did: $python setup.py build but got the following errors: uild/temp.linux-i686-2.3/Lib/xplt/src/play/x11/colors.o -DGISTPATH="/usr/lib/python2.3/site-packages/scipy/xplt/gistdata" In file included from Lib/xplt/src/play/x11/colors.c:9: Lib/xplt/src/play/x11/playx.h:11:22: X11/Xlib.h: No such file or directory In file included from Lib/xplt/src/play/x11/colors.c:9: Lib/xplt/src/play/x11/playx.h:29: error: parse error before "Display" Lib/xplt/src/play/x11/playx.h:29: warning: no semicolon at end of struct or union Lib/xplt/src/play/x11/playx.h:31: warning: type defaults to `int' in declaration of `wm_protocols' Lib/xplt/src/play/x11/playx.h:31: warning: type defaults to `int' in declaration of `wm_delete' Lib/xplt/src/play/x11/playx.h:31: warning: data definition has no type or storage class Lib/xplt/src/play/x11/playx.h:34: error: parse error before '*' token Lib/xplt/src/play/x11/playx.h:34: warning: type defaults to `int' in declaration of `font' Lib/xplt/src/play/x11/playx.h:34: warning: data definition has no type or storage class Lib/xplt/src/play/x11/playx.h:38: error: parse error before "XFontStruct" Lib/xplt/src/play/x11/playx.h:38: warning: no semicolon at end of struct or union Lib/xplt/src/play/x11/playx.h:39: error: conflicting types for `font' Lib/xplt/src/play/x11/playx.h:34: error: previous declaration of `font' Lib/xplt/src/play/x11/playx.h:40: error: parse error before '}' token Lib/xplt/src/play/x11/playx.h:40: warning: type defaults to `int' in declaration of `cached' Lib/xplt/src/play/x11/playx.h:40: warning: data definition has no type or storage class Lib/xplt/src/play/x11/playx.h:48: error: parse error before "cursors" Lib/xplt/src/play/x11/playx.h:48: warning: type defaults to `int' in declaration of `cursors' Lib/xplt/src/play/x11/playx.h:48: warning: data definition has no type or storage class Lib/xplt/src/play/x11/playx.h:62: error: parse error before '}' token Lib/xplt/src/play/x11/playx.h:72: error: parse error before "Window" Lib/xplt/src/play/x11/playx.h:72: warning: no semicolon at end of struct or union Lib/xplt/src/play/x11/playx.h:81: error: parse error before "cmap" Lib/xplt/src/play/x11/playx.h:81: warning: type defaults to `int' in declaration of `cmap' Lib/xplt/src/play/x11/playx.h:81: warning: data definition has no type or storage class Lib/xplt/src/play/x11/playx.h:83: error: parse error before "colors" Lib/xplt/src/play/x11/playx.h:83: warning: type defaults to `int' in declaration of `colors' Lib/xplt/src/play/x11/playx.h:83: warning: data definition has no type or storage class Lib/xplt/src/play/x11/playx.h:85: error: parse error before "gray" Lib/xplt/src/play/x11/playx.h:85: warning: type defaults to `int' in declaration of `gray' Lib/xplt/src/play/x11/playx.h:85: warning: data definition has no type or storage class Lib/xplt/src/play/x11/playx.h:90: error: parse error before "gc" Lib/xplt/src/play/x11/playx.h:90: warning: type defaults to `int' in declaration of `gc' Lib/xplt/src/play/x11/playx.h:90: warning: data definition has no type or storage class Lib/xplt/src/play/x11/playx.h:99: error: parse error before '*' token Lib/xplt/src/play/x11/playx.h:99: warning: type defaults to `int' in declaration of `image' Lib/xplt/src/play/x11/playx.h:99: warning: data definition has no type or storage class Lib/xplt/src/play/x11/playx.h:101: error: parse error before "pixmap" Lib/xplt/src/play/x11/playx.h:101: warning: type defaults to `int' in declaration of `pixmap' Lib/xplt/src/play/x11/playx.h:101: warning: data definition has no type or storage class Lib/xplt/src/play/x11/playx.h:102: error: parse error before "rotgc" Lib/xplt/src/play/x11/playx.h:102: warning: type defaults to `int' in declaration of `rotgc' Lib/xplt/src/play/x11/playx.h:102: warning: data definition has no type or storage class Lib/xplt/src/play/x11/playx.h:104: error: parse error before '}' token Lib/xplt/src/play/x11/playx.h:110: error: parse error before "Drawable" Lib/xplt/src/play/x11/playx.h:110: warning: no semicolon at end of struct or union Lib/xplt/src/play/x11/playx.h:114: error: parse error before "cmap" Lib/xplt/src/play/x11/playx.h:114: warning: type defaults to `int' in declaration of `cmap' Lib/xplt/src/play/x11/playx.h:114: warning: data definition has no type or storage class Lib/xplt/src/play/x11/playx.h:118: error: parse error before '}' token Lib/xplt/src/play/x11/playx.h:121: error: parse error before "x_pt_list" Lib/xplt/src/play/x11/playx.h:121: warning: type defaults to `int' in declaration of `x_pt_list' Lib/xplt/src/play/x11/playx.h:121: warning: data definition has no type or storage class Lib/xplt/src/play/x11/playx.h:127: error: parse error before '*' token Lib/xplt/src/play/x11/playx.h:127: warning: function declaration isn't a prototype Lib/xplt/src/play/x11/playx.h:128: error: parse error before "Drawable" Lib/xplt/src/play/x11/playx.h:128: warning: function declaration isn't a prototype Lib/xplt/src/play/x11/playx.h:131: error: parse error before '*' token Lib/xplt/src/play/x11/playx.h:131: warning: function declaration isn't a prototype Lib/xplt/src/play/x11/playx.h:132: error: parse error before '*' token Lib/xplt/src/play/x11/playx.h:132: warning: function declaration isn't a prototype Lib/xplt/src/play/x11/playx.h:142: error: parse error before '*' token Lib/xplt/src/play/x11/playx.h:142: warning: type defaults to `int' in declaration of `x_font' Lib/xplt/src/play/x11/playx.h:142: warning: data definition has no type or storage class Lib/xplt/src/play/x11/playx.h:143: error: parse error before '*' token Lib/xplt/src/play/x11/playx.h:143: warning: function declaration isn't a prototype Lib/xplt/src/play/x11/playx.h:144: error: parse error before "x_getgc" Lib/xplt/src/play/x11/playx.h:144: warning: type defaults to `int' in declaration of `x_getgc' Lib/xplt/src/play/x11/playx.h:144: warning: data definition has no type or storage class Lib/xplt/src/play/x11/playx.h:157: error: parse error before "x_cursor" Lib/xplt/src/play/x11/playx.h:157: warning: type defaults to `int' in declaration of `x_cursor' Lib/xplt/src/play/x11/playx.h:157: warning: data definition has no type or storage class Lib/xplt/src/play/x11/playx.h:162: error: parse error before '*' token Lib/xplt/src/play/x11/playx.h:162: warning: function declaration isn't a prototype Lib/xplt/src/play/x11/playx.h:164: error: parse error before '*' token Lib/xplt/src/play/x11/playx.h:164: warning: function declaration isn't a prototype Lib/xplt/src/play/x11/playx.h:165: error: parse error before '*' token Lib/xplt/src/play/x11/playx.h:165: warning: function declaration isn't a prototype Lib/xplt/src/play/x11/colors.c:12: error: parse error before "x_getgc" Lib/xplt/src/play/x11/colors.c:13: warning: return type defaults to `int' Lib/xplt/src/play/x11/colors.c: In function `x_getgc': Lib/xplt/src/play/x11/colors.c:14: error: `GC' undeclared (first use in this function) Lib/xplt/src/play/x11/colors.c:14: error: (Each undeclared identifier is reported only once Lib/xplt/src/play/x11/colors.c:14: error: for each function it appears in.) Lib/xplt/src/play/x11/colors.c:14: error: parse error before "gc" Lib/xplt/src/play/x11/colors.c:15: error: dereferencing pointer to incomplete type Lib/xplt/src/play/x11/colors.c:16: error: dereferencing pointer to incomplete type Lib/xplt/src/play/x11/colors.c:17: error: dereferencing pointer to incomplete type Lib/xplt/src/play/x11/colors.c:17: error: dereferencing pointer to incomplete type Lib/xplt/src/play/x11/colors.c:17: error: dereferencing pointer to incomplete type Lib/xplt/src/play/x11/colors.c:17: error: dereferencing pointer to incomplete type Lib/xplt/src/play/x11/colors.c:18: error: dereferencing pointer to incomplete type Lib/xplt/src/play/x11/colors.c:20: error: dereferencing pointer to incomplete type Lib/xplt/src/play/x11/colors.c:26: warning: implicit declaration of function `XSetFillStyle' Lib/xplt/src/play/x11/colors.c:26: error: dereferencing pointer to incomplete type Lib/xplt/src/play/x11/colors.c:27: error: dereferencing pointer to incomplete type Lib/xplt/src/play/x11/colors.c: In function `p_color': Lib/xplt/src/play/x11/colors.c:35: error: dereferencing pointer to incomplete type Lib/xplt/src/play/x11/colors.c:36: error: `GC' undeclared (first use in this function) Lib/xplt/src/play/x11/colors.c:36: error: parse error before "gc" Lib/xplt/src/play/x11/colors.c:37: error: dereferencing pointer to incomplete type Lib/xplt/src/play/x11/colors.c:42: error: dereferencing pointer to incomplete type Lib/xplt/src/play/x11/colors.c:45: warning: implicit declaration of function `XSetFunction' Lib/xplt/src/play/x11/colors.c:45: error: dereferencing pointer to incomplete type Lib/xplt/src/play/x11/colors.c:45: error: `GXxor' undeclared (first use in this function) Lib/xplt/src/play/x11/colors.c:47: error: dereferencing pointer to incomplete type Lib/xplt/src/play/x11/colors.c:47: error: `GXcopy' undeclared (first use in this function) Lib/xplt/src/play/x11/colors.c:49: error: dereferencing pointer to incomplete type Lib/xplt/src/play/x11/colors.c:50: error: dereferencing pointer to incomplete type Lib/xplt/src/play/x11/colors.c:50: error: `FillOpaqueStippled' undeclared (first use in this function) Lib/xplt/src/play/x11/colors.c:51: warning: implicit declaration of function `XSetStipple' Lib/xplt/src/play/x11/colors.c:51: error: dereferencing pointer to incomplete type Lib/xplt/src/play/x11/colors.c:51: error: dereferencing pointer to incomplete type Lib/xplt/src/play/x11/colors.c:52: warning: implicit declaration of function `XSetBackground' Lib/xplt/src/play/x11/colors.c:52: error: dereferencing pointer to incomplete type Lib/xplt/src/play/x11/colors.c:52: error: dereferencing pointer to incomplete type Lib/xplt/src/play/x11/colors.c:54: error: dereferencing pointer to incomplete type Lib/xplt/src/play/x11/colors.c:55: error: dereferencing pointer to incomplete type Lib/xplt/src/play/x11/colors.c:55: error: `FillSolid' undeclared (first use in this function) Lib/xplt/src/play/x11/colors.c:56: error: dereferencing pointer to incomplete type Lib/xplt/src/play/x11/colors.c:56: error: dereferencing pointer to incomplete type Lib/xplt/src/play/x11/colors.c:59: warning: implicit declaration of function `XSetForeground' Lib/xplt/src/play/x11/colors.c:59: error: dereferencing pointer to incomplete type Lib/xplt/src/play/x11/colors.c:60: error: dereferencing pointer to incomplete type Lib/xplt/src/play/x11/colors.c: In function `x_getpixel': Lib/xplt/src/play/x11/colors.c:67: error: dereferencing pointer to incomplete type Lib/xplt/src/play/x11/colors.c:69: error: dereferencing pointer to incomplete type Lib/xplt/src/play/x11/colors.c:69: error: dereferencing pointer to incomplete type Lib/xplt/src/play/x11/colors.c:71: error: dereferencing pointer to incomplete type Lib/xplt/src/play/x11/colors.c:77: error: dereferencing pointer to incomplete type Lib/xplt/src/play/x11/colors.c:77: error: `TrueColor' undeclared (first use in this function) Lib/xplt/src/play/x11/colors.c:77: error: dereferencing pointer to incomplete type Lib/xplt/src/play/x11/colors.c:77: error: `DirectColor' undeclared (first use in this function) Lib/xplt/src/play/x11/colors.c:78: error: dereferencing pointer to incomplete type Lib/xplt/src/play/x11/colors.c:78: error: dereferencing pointer to incomplete type Lib/xplt/src/play/x11/colors.c:79: error: dereferencing pointer to incomplete type Lib/xplt/src/play/x11/colors.c:79: error: dereferencing pointer to incomplete type Lib/xplt/src/play/x11/colors.c:79: error: dereferencing pointer to incomplete type Lib/xplt/src/play/x11/colors.c:79: error: dereferencing pointer to incomplete type Lib/xplt/src/play/x11/colors.c:80: error: dereferencing pointer to incomplete type Lib/xplt/src/play/x11/colors.c:80: error: `PseudoColor' undeclared (first use in this function) Lib/xplt/src/play/x11/colors.c:81: error: dereferencing pointer to incomplete type Lib/xplt/src/play/x11/colors.c:82: error: dereferencing pointer to incomplete type Lib/xplt/src/play/x11/colors.c:88: error: dereferencing pointer to incomplete type Lib/xplt/src/play/x11/colors.c:90: error: dereferencing pointer to incomplete type error: command 'gcc' failed with exit status 1 shaman at peyote:/usr/lib/python2.3/site-packages/scipy$ From pearu at scipy.org Tue Jan 27 02:35:02 2004 From: pearu at scipy.org (Pearu Peterson) Date: Tue, 27 Jan 2004 01:35:02 -0600 (CST) Subject: [SciPy-user] need help with gcc error In-Reply-To: <601d060407.60407601d0@ucmerced.edu> Message-ID: On Mon, 26 Jan 2004 ehaux at ucmerced.edu wrote: > can anyone help with this?? > i downloaded from cvs and did: > $python setup.py build > > but got the following errors: > > uild/temp.linux-i686-2.3/Lib/xplt/src/play/x11/colors.o -DGISTPATH="/usr/lib/python2.3/site-packages/scipy/xplt/gistdata" > In file included from Lib/xplt/src/play/x11/colors.c:9: > Lib/xplt/src/play/x11/playx.h:11:22: X11/Xlib.h: No such file or directory ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ What is the output of python scipy_core/scipy_distutils/system_info.py ? Where the header file Xlib.h is located in your system? And could you give more details about your system (sys.platform,os.name, uname -a, etc)? Pearu From gazzar at email.com Tue Jan 27 03:46:17 2004 From: gazzar at email.com (Gary Ruben) Date: Tue, 27 Jan 2004 18:46:17 +1000 Subject: [SciPy-user] Noticed scipy for py2.3 binary, but have a problem Message-ID: <20040127084617.25103.qmail@email.com> I've been waiting impatiently for months for a Windows binary installer for python 2.3 to appear and I see one has very quietly snuck onto the download page. Woohoo! methinks. However, at first attempt, I get >>> from scipy import * Traceback (most recent call last): File "", line 1, in -toplevel- import scipy File "C:\APPS\PYTHON23\Lib\site-packages\scipy\__init__.py", line 11, in -toplevel- from scipy_base import * File "C:\APPS\PYTHON23\lib\site-packages\scipy_base\__init__.py", line 117, in -toplevel- import fastumath # no need to use scipy_base.fastumath ImportError: DLL load failed: One of the library files needed to run this application cannot be found. I'm running Win98, Python 2.3.3 (and will be putting it on my Win2000 machine when I get it working). I notice that there's a fastumath.pyd in scipy_base, but should there also be a .dll file in here? In fact, there are no dlls in any of the scipy directories. Gary Ruben -- ___________________________________________________________ Sign-up for Ads Free at Mail.com http://promo.mail.com/adsfreejump.htm From scipy at zunzun.com Tue Jan 27 06:17:31 2004 From: scipy at zunzun.com (James R. Phillips) Date: Tue, 27 Jan 2004 06:17:31 -500 Subject: [SciPy-user] need help with gcc error References: <601d060407.60407601d0@ucmerced.edu> Message-ID: <401648cb9456c1.76981309@mercury.sabren.com> > i downloaded from cvs and did $python setup.py build > but got the following errors: > > Lib/xplt/src/play/x11/playx.h:11:22: X11/Xlib.h: No such file or directory You may need to configure the setup.py file for scipy to not build anything with X windows. Here is the file I used as an example: http://zunzun.com/no_X_setup.py James Phillips From josh at cs.byu.edu Tue Jan 27 12:11:35 2004 From: josh at cs.byu.edu (Joshua Menke) Date: Tue, 27 Jan 2004 10:11:35 -0700 Subject: [SciPy-user] Running more than one applications using weave Message-ID: <40169BC7.3080500@cs.byu.edu> I'm having problems when I run more than one app. using weave on the same (dual-processor) computer. Both processes will often attempt to create / delete / access the directories weave uses to store the compiled code in at the same time, causing one of the applications to fail. Is there a way to force weave to only use the local / current directory for *all* its files? I know how to tell it which directory to prefer using the PYTHONCOMPILED environment variables, but weave still uses other directories for intermediate files (/tmp/%USER%/) and also always tries to create and verify the default directory ~/.python_compiled. Is there a way to force it NOT to use anywhere but the current directory? Or is there another way to fix this problem? I'd like to be able to use a cluster of dual-procs machines we have, but the same problem would occur on the machines. In case you need more specific info on the most common error I see, it happens when weave tries to verify a directory is writable. My first application will create a dummy file then the second app will create one, the first app will delete the dummy file, then the second app will fail to delete the now non-existent dummy file, and the program terminates on the uncaught exception. thanks, Josh -- Joshua Menke NNML Lab BYU CS Dept. josh at cs.byu.edu From chris at fonnesbeck.org Tue Jan 27 15:07:45 2004 From: chris at fonnesbeck.org (Christopher Fonnesbeck) Date: Tue, 27 Jan 2004 15:07:45 -0500 Subject: [SciPy-user] still cant save plots in OSX Message-ID: <78E64996-5104-11D8-B569-000A956FDAC0@fonnesbeck.org> Months on, I am still unable to successfully save plots from scipy.gplt to files . I get no messages or errors, and most importantly, no png or jpg file. Is there any way of getting better feedback from Scipy so that I can debug this? Using recent CVS version on OSX 10.3 Thanks, Chris -- Christopher J. Fonnesbeck ( c h r i s @ f o n n e s b e c k . o r g ) Georgia Cooperative Fish & Wildlife Research Unit, University of Georgia From chris at fonnesbeck.org Tue Jan 27 17:06:31 2004 From: chris at fonnesbeck.org (Christopher Fonnesbeck) Date: Tue, 27 Jan 2004 17:06:31 -0500 Subject: [SciPy-user] still cant save plots in OSX In-Reply-To: <78E64996-5104-11D8-B569-000A956FDAC0@fonnesbeck.org> References: <78E64996-5104-11D8-B569-000A956FDAC0@fonnesbeck.org> Message-ID: <10079B6A-5115-11D8-A3CE-000A956FDAC0@fonnesbeck.org> On Jan 27, 2004, at 3:07 PM, Christopher Fonnesbeck wrote: > Months on, I am still unable to successfully save plots from > scipy.gplt to files . I get no messages or errors, and most > importantly, no png or jpg file. Is there any way of getting better > feedback from Scipy so that I can debug this? > > Using recent CVS version on OSX 10.3 > An interesting result, following more investigation: when scipy.gplt.output(filename,'png color') is called, no file is written. However, when the second argument is changed to simply 'png', the file is written, and in color! No idea why this works -- perhaps someone with more gnuplot experience can shed light on it. Chris -- Christopher J. Fonnesbeck ( c h r i s @ f o n n e s b e c k . o r g ) Georgia Cooperative Fish & Wildlife Research Unit, University of Georgia From nwagner at mecha.uni-stuttgart.de Thu Jan 29 07:07:54 2004 From: nwagner at mecha.uni-stuttgart.de (Nils Wagner) Date: Thu, 29 Jan 2004 13:07:54 +0100 Subject: [SciPy-user] Hessenberg form of a matrix Message-ID: <4018F79A.9BB481F9@mecha.uni-stuttgart.de> Dear experts, Has some written a function for scipy which produces a Hessenberg matrix ? This task is solved by *GEHRD in LAPACK. I am grateful for any hint. Nils From arnd.baecker at web.de Thu Jan 29 08:25:16 2004 From: arnd.baecker at web.de (Arnd Baecker) Date: Thu, 29 Jan 2004 14:25:16 +0100 (CET) Subject: [SciPy-user] Hessenberg form of a matrix In-Reply-To: <4018F79A.9BB481F9@mecha.uni-stuttgart.de> References: <4018F79A.9BB481F9@mecha.uni-stuttgart.de> Message-ID: Hi Nils, On Thu, 29 Jan 2004, Nils Wagner wrote: > Has some written a function for scipy which produces a Hessenberg matrix > ? > This task is solved by > > *GEHRD > > in LAPACK. Nothing working seems to be available. So I recommend to use f2py2e, http://cens.ioc.ee/projects/f2py2e/ *GEHRD are not too complicated at first glance, so this will be pretty straightforward with the good documentation. Give it a try and if you get stuck send me the status+files (off-list) and I can have a look. I think it would be great if we can get more and more routines of this type included in scipy by user contributions... Best, Arnd From chris at fonnesbeck.org Thu Jan 29 10:36:38 2004 From: chris at fonnesbeck.org (Christopher Fonnesbeck) Date: Thu, 29 Jan 2004 10:36:38 -0500 Subject: [SciPy-user] Making faster statistical distributions Message-ID: I am already using pieces of SciPy in my Markov chain Monte Carlo package (PyMC), mostly for plotting functionality. I would also like to exploit the distributions implemented in scipy.stats, but they are far too slow for use in statistical simulation applications like MCMC, where millions of random draws may be taken. Therefore, I am thinking of implementing many of these distributions (at least the common ones) as C or Fortran extensions. I am unsure whether to use Fortran through f2py for this task, or C through weave.inline (for example). I have used both in the past for various tasks, and was generally happy with both. Any suggestions? Thanks, C. -- Christopher J. Fonnesbeck ( c h r i s @ f o n n e s b e c k . o r g ) Georgia Cooperative Fish & Wildlife Research Unit, University of Georgia From josh at cs.byu.edu Thu Jan 29 10:41:31 2004 From: josh at cs.byu.edu (Joshua Menke) Date: Thu, 29 Jan 2004 08:41:31 -0700 Subject: [SciPy-user] Running more than one applications using weave doesn't work well In-Reply-To: <40169BC7.3080500@cs.byu.edu> References: <40169BC7.3080500@cs.byu.edu> Message-ID: <401929AB.3000505@cs.byu.edu> I'm guessing no one knows the answer to this one, or it's too obvious? Should I post this on the dev list instead? It's a serious concern for me thanks, Josh Joshua Menke wrote: > I'm having problems when I run more than one app. using weave on the > same (dual-processor) computer. Both processes will often attempt to > create / delete / access the directories weave uses to store the > compiled code in at the same time, causing one of the applications to > fail. Is there a way to force weave to only use the local / current > directory for *all* its files? I know how to tell it which directory > to prefer using the PYTHONCOMPILED environment variables, but weave > still uses other directories for intermediate files (/tmp/%USER%/) and > also always tries to create and verify the default directory > ~/.python_compiled. Is there a way to force it NOT to use anywhere but > the current directory? Or is there another way to fix this problem? > I'd like to be able to use a cluster of dual-procs machines we have, > but the same problem would occur on the machines. > > In case you need more specific info on the most common error I see, it > happens when weave tries to verify a directory is writable. My first > application will create a dummy file then the second app will create > one, the first app will delete the dummy file, then the second app > will fail to delete the now non-existent dummy file, and the program > terminates on the uncaught exception. > > thanks, > > Josh > -- Joshua Menke NNML Lab BYU CS josh at cs.byu.edu From jdhunter at ace.bsd.uchicago.edu Thu Jan 29 10:54:44 2004 From: jdhunter at ace.bsd.uchicago.edu (John Hunter) Date: Thu, 29 Jan 2004 09:54:44 -0600 Subject: [SciPy-user] Making faster statistical distributions In-Reply-To: (Christopher Fonnesbeck's message of "Thu, 29 Jan 2004 10:36:38 -0500") References: Message-ID: >>>>> "Christopher" == Christopher Fonnesbeck writes: Christopher> I am already using pieces of SciPy in my Markov chain Christopher> Monte Carlo package (PyMC), mostly for plotting Christopher> functionality. I would also like to exploit the Christopher> distributions implemented in scipy.stats, but they Christopher> are far too slow for use in statistical simulation Christopher> applications like MCMC, where millions of random Christopher> draws may be taken. Therefore, I am thinking of Christopher> implementing many of these distributions (at least Christopher> the common ones) as C or Fortran extensions. I am Christopher> unsure whether to use Fortran through f2py for this Christopher> task, or C through weave.inline (for example). I have Christopher> used both in the past for various tasks, and was Christopher> generally happy with both. Any suggestions? I've wrote some code to fill Numeric arrays with levy distributions from gsl. GSL has many, many distributions - http://www.gnu.org/software/gsl/manual/gsl-ref_19.html#SEC284 Here's a sample test script from levy import randlevy, seed gamma = 0.003501 alpha = 1.341702 seed(1235) x = randlevy(100000, alpha, gamma) print min(x), max(x) Seeding is optional - the rng is seeded by the clock if you don't explicitly call seed. Below is the module code which you can use as a template if you're interested. I think it would be nice to extend this code to include all the gsl distributions, but haven't taken the time. pygsl wraps these distributions, but suffers from the repeated call overhead that is hurting you. It would be a natural extension of pygsl to include initializing numeric arrays from the distributions. #include "Python.h" #include "Numeric/arrayobject.h" #include #include #include #include #include static PyObject *ErrorObject; // get x[i] from a 1d array of type xtype #define get1d(x,i,xtype) \ *(xtype *)(x->data+i*x->strides[0]) int _levymodRNGSeeded = 0; gsl_rng * _levymodRNG; static PyObject * seed(PyObject *self, PyObject *args) { const gsl_rng_type * T; int N; T = gsl_rng_default; _levymodRNG = gsl_rng_alloc(T); gsl_rng_env_setup(); if (args==Py_None) { //printf("Default seed\n"); time_t t1; (void) time(&t1); srand48((long) t1); N = (int)t1; } else if (!PyArg_ParseTuple(args, "i", &N)) return NULL; //printf("Setting seed %d\n", N); gsl_rng_set (_levymodRNG, N); _levymodRNGSeeded = 1; Py_INCREF(Py_None); return Py_None; } static PyObject * randlevy(PyObject *self, PyObject *args) { if (_levymodRNGSeeded == 0) { seed(self, Py_None); } int dimensions[1]; int i; PyArrayObject *x; int N; float alpha; float gamma; if (!PyArg_ParseTuple(args, "iff", &N, &alpha, &gamma)) return NULL; dimensions[0] = N; x = (PyArrayObject *)PyArray_FromDims(1,dimensions,PyArray_FLOAT); for (i=0; i References: Message-ID: <40195C41.3000707@enthought.com> Christopher Fonnesbeck wrote: > I am already using pieces of SciPy in my Markov chain Monte Carlo > package (PyMC), mostly for plotting functionality. I would also like > to exploit the distributions implemented in scipy.stats, but they are > far too slow for use in statistical simulation applications like MCMC, > where millions of random draws may be taken. Therefore, I am thinking > of implementing many of these distributions (at least the common ones) > as C or Fortran extensions. I am unsure whether to use Fortran through > f2py for this task, or C through weave.inline (for example). I have > used both in the past for various tasks, and was generally happy with > both. Any suggestions? Could you specify which ones are too slow? This is a rather broad statement as many are implemented in C and are very fast. Some distributions, however, do default to using a numerical solver to invert the cdf and apply this to uniform random variates. You can improve the speed of these distributions by overriding the _ppf method or the _rvs method of the object to use a faster, more specialized method. I would use weave or fortran with f2py to do this. Best, -Travis O. > > Thanks, > C. > -- > Christopher J. Fonnesbeck ( c h r i s @ f o n n e s b e c k . o r g ) > Georgia Cooperative Fish & Wildlife Research Unit, University of Georgia > > _______________________________________________ > SciPy-user mailing list > SciPy-user at scipy.net > http://www.scipy.net/mailman/listinfo/scipy-user From chris at fonnesbeck.org Thu Jan 29 15:53:10 2004 From: chris at fonnesbeck.org (Christopher Fonnesbeck) Date: Thu, 29 Jan 2004 15:53:10 -0500 Subject: [SciPy-user] Making faster statistical distributions In-Reply-To: <40195C41.3000707@enthought.com> References: <40195C41.3000707@enthought.com> Message-ID: <25DB45DA-529D-11D8-A1E7-000A956FDAC0@fonnesbeck.org> On Jan 29, 2004, at 2:17 PM, Travis Oliphant wrote: > Christopher Fonnesbeck wrote: > >> I am already using pieces of SciPy in my Markov chain Monte Carlo >> package (PyMC), mostly for plotting functionality. I would also like >> to exploit the distributions implemented in scipy.stats, but they are >> far too slow for use in statistical simulation applications like >> MCMC, where millions of random draws may be taken. Therefore, I am >> thinking of implementing many of these distributions (at least the >> common ones) as C or Fortran extensions. I am unsure whether to use >> Fortran through f2py for this task, or C through weave.inline (for >> example). I have used both in the past for various tasks, and was >> generally happy with both. Any suggestions? > > > Could you specify which ones are too slow? This is a rather broad > statement as many are implemented in C and are very fast. Some > distributions, however, do default to using a numerical solver to > invert the cdf and apply this to uniform random variates. You can > improve the speed of these distributions by overriding the _ppf > method or the _rvs method of the object to use a faster, more > specialized method. I would use weave or fortran with f2py to do > this. > > Best, > > -Travis O. Well, the binomial and normal distributions, for sure, off the top of my head. Using the scipy distributions slows my MCMC code down significantly (they were the bottleneck, according to the profiling module). Using Fortran via f2py sped things up a lot. I'm not talking about the generation of random deviates, necessarily, but rather the pdf's, which are used for calculating likelihoods. C. -- Christopher J. Fonnesbeck ( c h r i s @ f o n n e s b e c k . o r g ) Georgia Cooperative Fish & Wildlife Research Unit, University of Georgia From rkern at ucsd.edu Thu Jan 29 16:28:47 2004 From: rkern at ucsd.edu (Robert Kern) Date: Thu, 29 Jan 2004 13:28:47 -0800 Subject: [SciPy-user] Making faster statistical distributions In-Reply-To: <25DB45DA-529D-11D8-A1E7-000A956FDAC0@fonnesbeck.org> References: <40195C41.3000707@enthought.com> <25DB45DA-529D-11D8-A1E7-000A956FDAC0@fonnesbeck.org> Message-ID: <20040129212847.GA10178@taliesen.ucsd.edu> On Thu, Jan 29, 2004 at 03:53:10PM -0500, Christopher Fonnesbeck wrote: [snip] > Well, the binomial and normal distributions, for sure, off the top of > my head. Using the scipy distributions slows my MCMC code down > significantly (they were the bottleneck, according to the profiling > module). Using Fortran via f2py sped things up a lot. I'm not talking > about the generation of random deviates, necessarily, but rather the > pdf's, which are used for calculating likelihoods. I'm willing to bet that the speed is lost by the generality of the framework. There's a lot of argument manipulation and not a lot of actual computation. It might be useful to add a fastdists module that computes pdfs and cdfs in a more specific (but fast) way for the more common distributions. And log-pdfs, too, please! In fact, log-pdfs first since in many instances like MCMC, they're much more suitable. One could then override pdf() and cdf() (not _pdf() and _cdf()) in the specific classes. But that's a design that appears to have been decided against. > C. -- Robert Kern rkern at ucsd.edu "In the fields of hell where the grass grows high Are the graves of dreams allowed to die." -- Richard Harter From oliphant at ee.byu.edu Thu Jan 29 20:17:45 2004 From: oliphant at ee.byu.edu (Travis Oliphant) Date: Thu, 29 Jan 2004 18:17:45 -0700 Subject: [SciPy-user] Making faster statistical distributions In-Reply-To: <20040129212847.GA10178@taliesen.ucsd.edu> References: <40195C41.3000707@enthought.com> <25DB45DA-529D-11D8-A1E7-000A956FDAC0@fonnesbeck.org> <20040129212847.GA10178@taliesen.ucsd.edu> Message-ID: <4019B0B9.1090601@ee.byu.edu> Robert Kern wrote: >On Thu, Jan 29, 2004 at 03:53:10PM -0500, Christopher Fonnesbeck wrote: > >[snip] > > > >>Well, the binomial and normal distributions, for sure, off the top of >>my head. Using the scipy distributions slows my MCMC code down >>significantly (they were the bottleneck, according to the profiling >>module). Using Fortran via f2py sped things up a lot. I'm not talking >>about the generation of random deviates, necessarily, but rather the >>pdf's, which are used for calculating likelihoods. >> >> > >I'm willing to bet that the speed is lost by the generality of the >framework. There's a lot of argument manipulation and not a lot of >actual computation. It might be useful to add a fastdists module that >computes pdfs and cdfs in a more specific (but fast) way for the more >common distributions. And log-pdfs, too, please! In fact, log-pdfs >first since in many instances like MCMC, they're much more suitable. > >One could then override pdf() and cdf() (not _pdf() and _cdf()) in the >specific classes. But that's a design that appears to have been decided >against. > > I wouldn't say anything has been decided against at this point. We could pretty easily over-ride pdf and cdf for whatevern distribution you like. Perhaps it would be a good idea to do so for the most common pdfs. -Travis O. From gazzar at email.com Thu Jan 29 19:28:02 2004 From: gazzar at email.com (Gary Ruben) Date: Fri, 30 Jan 2004 10:28:02 +1000 Subject: [SciPy-user] Noticed scipy for py2.3 binary, but have a problem Message-ID: <20040130002802.1203.qmail@email.com> For anyone reading this thread, Travis sent me the scipy-core installer and it appears to have worked and gotten me up and running. Thanks Travis. ----- Original Message ----- From: "Gary Ruben" Date: Tue, 27 Jan 2004 18:46:17 +1000 To: scipy-user at scipy.net Subject: [SciPy-user] Noticed scipy for py2.3 binary, but have a problem > I've been waiting impatiently for months for a Windows binary installer for python 2.3 to appear and I see one has very quietly snuck onto the download page. Woohoo! methinks. However, at first attempt, I get > > >>> from scipy import * > > Traceback (most recent call last): > File "", line 1, in -toplevel- > import scipy > File "C:\APPS\PYTHON23\Lib\site-packages\scipy\__init__.py", line 11, in -toplevel- > from scipy_base import * > File "C:\APPS\PYTHON23\lib\site-packages\scipy_base\__init__.py", line 117, in -toplevel- > import fastumath # no need to use scipy_base.fastumath > ImportError: DLL load failed: One of the library files needed to run this application cannot be found. > > I'm running Win98, Python 2.3.3 (and will be putting it on my Win2000 machine when I get it working). > I notice that there's a fastumath.pyd in scipy_base, but should there also be a .dll file in here? In fact, there are no dlls in any of the scipy directories. > > Gary Ruben > -- -- ___________________________________________________________ Sign-up for Ads Free at Mail.com http://promo.mail.com/adsfreejump.htm From chris at fonnesbeck.org Thu Jan 29 20:05:41 2004 From: chris at fonnesbeck.org (Christopher Fonnesbeck) Date: Thu, 29 Jan 2004 20:05:41 -0500 Subject: [SciPy-user] Making faster statistical distributions In-Reply-To: <4019B0B9.1090601@ee.byu.edu> References: <40195C41.3000707@enthought.com> <25DB45DA-529D-11D8-A1E7-000A956FDAC0@fonnesbeck.org> <4019B0B9.1090601@ee.byu.edu> Message-ID: <6C56CD68-52C0-11D8-95B1-000A956FDAC0@fonnesbeck.org> On Jan 29, 2004, at 8:17 PM, Travis Oliphant wrote: > Robert Kern wrote: > >> On Thu, Jan 29, 2004 at 03:53:10PM -0500, Christopher Fonnesbeck >> wrote: >> >> [snip] >> >> >>> Well, the binomial and normal distributions, for sure, off the top >>> of my head. Using the scipy distributions slows my MCMC code down >>> significantly (they were the bottleneck, according to the profiling >>> module). Using Fortran via f2py sped things up a lot. I'm not >>> talking about the generation of random deviates, necessarily, but >>> rather the pdf's, which are used for calculating likelihoods. >>> >> >> I'm willing to bet that the speed is lost by the generality of the >> framework. There's a lot of argument manipulation and not a lot of >> actual computation. It might be useful to add a fastdists module that >> computes pdfs and cdfs in a more specific (but fast) way for the more >> common distributions. And log-pdfs, too, please! In fact, log-pdfs >> first since in many instances like MCMC, they're much more suitable. >> >> One could then override pdf() and cdf() (not _pdf() and _cdf()) in the >> specific classes. But that's a design that appears to have been >> decided >> against. >> > > I wouldn't say anything has been decided against at this point. We > could pretty easily over-ride pdf and cdf for whatevern distribution > you like. Perhaps it would be a good idea to do so for the most > common pdfs. > -Travis O. Perhaps I will take a shot at this, then, since I already have several Fortran extensions written for log-likelihoods. C. -- Christopher J. Fonnesbeck ( c h r i s @ f o n n e s b e c k . o r g ) Georgia Cooperative Fish & Wildlife Research Unit, University of Georgia From adrian.blockley at environment.wa.gov.au Fri Jan 30 02:13:13 2004 From: adrian.blockley at environment.wa.gov.au (BLOCKLEY Adrian) Date: Fri, 30 Jan 2004 15:13:13 +0800 Subject: [SciPy-user] Installing under Windows - what am I doing wrong? Message-ID: <04Jan30.152623wst.118089@epagate.environ.wa.gov.au> Hi All, After successfully installing SciPi under Linux I decided to install under Windows as I am investigating using SciPi for some cross platform software that we are developing. The installation seems to have worked but when I tried running some tests I get an error. Traceback (most recent call last): File "J:\temp\PLATLIB\scipy\tests\test_basic.py", line 6, in ? import scipy_base.limits as limits ImportError: No module named scipy_base.limits Now I confess to being a total Python Newbie but it seems to me that I need a module named scipy_base but I cannot find anything like this. Any suggestions as to what I am doing wrong? I am running Python 2.2 under windows 2000. I installed Scipy using SciPy-0.2.1_253.3691.win32-py2.2.exe I've notice under the Linux install that there is a whole directory called scipy_base but this seems to be missing in my windows install. Thanks Adrian ======================================================================= Adrian Blockley Western Australian Department of Environment 123 Adelaide Tce East Perth Mailing address: PO Box K822 Perth WA 6842 Ph: +61 8 9278 0692 Fax: +61 8 9278 0639 ======================================================================= Disclaimer: This e-mail is confidential to the addressee and is the view of the writer, not necessarily that of the Department of Environmental Protection, which accepts no responsibility for the contents. If you are not the addressee, please notify the Department by return e-mail and delete the message from your system. You must not disclose or use this information in any way. No warranty is made that this material is free from computer viruses. From jdhunter at ace.bsd.uchicago.edu Fri Jan 30 07:00:26 2004 From: jdhunter at ace.bsd.uchicago.edu (John Hunter) Date: Fri, 30 Jan 2004 06:00:26 -0600 Subject: [SciPy-user] Installing under Windows - what am I doing wrong? In-Reply-To: <04Jan30.152623wst.118089@epagate.environ.wa.gov.au> (BLOCKLEY Adrian's message of "Fri, 30 Jan 2004 15:13:13 +0800") References: <04Jan30.152623wst.118089@epagate.environ.wa.gov.au> Message-ID: >>>>> "BLOCKLEY" == BLOCKLEY Adrian writes: BLOCKLEY> Now I confess to being a total Python Newbie but it BLOCKLEY> seems to me that I need a module named scipy_base but I BLOCKLEY> cannot find anything like this. Any suggestions as to BLOCKLEY> what I am doing wrong? If it's an option for you, you may want to consider using the enthought edition of python for windows - http://enthought.com/python. Not only does this have scipy preinstalled, it comes with a lot of other useful packages for scientific computing and graphics. JDH From jose.martin at wanadoo.es Fri Jan 30 15:08:55 2004 From: jose.martin at wanadoo.es (=?iso-8859-1?Q?Jos=E9_A_Mart=EDn_H?=) Date: Fri, 30 Jan 2004 21:08:55 +0100 Subject: [SciPy-user] HELP !!! sci py dost not work !!! Message-ID: <002401c3e76c$e46ae200$52806750@CASA> IDLE 1.0.2 >>> import scipy Traceback (most recent call last): File "", line 1, in -toplevel- import scipy File "C:\Python23\Lib\site-packages\scipy\__init__.py", line 11, in -toplevel- from scipy_base import * ImportError: No module named scipy_base >>> i have upgraded to python 2.3 and downloaded the last binaries for scipy !!!. -------------- next part -------------- An HTML attachment was scrubbed... URL: From chris at fonnesbeck.org Fri Jan 30 15:05:32 2004 From: chris at fonnesbeck.org (Christopher Fonnesbeck) Date: Fri, 30 Jan 2004 15:05:32 -0500 Subject: [SciPy-user] HELP !!! sci py dost not work !!! In-Reply-To: <002401c3e76c$e46ae200$52806750@CASA> References: <002401c3e76c$e46ae200$52806750@CASA> Message-ID: On Jan 30, 2004, at 3:08 PM, Jos? A Mart?n H wrote: > IDLE 1.0.2????? > >>> import scipy > ? > Traceback (most recent call last): > ? File "", line 1, in -toplevel- > ??? import scipy > ? File "C:\Python23\Lib\site-packages\scipy\__init__.py", line 11, in > -toplevel- > ??? from scipy_base import * > ImportError: No module named scipy_base > >>> > ? > i have upgraded to python 2.3 and downloaded the last binaries for > scipy !!!. You are much better off compiling scipy from CVS than using the binaries. C. -- Christopher J. Fonnesbeck ( c h r i s @ f o n n e s b e c k . o r g ) Georgia Cooperative Fish & Wildlife Research Unit, University of Georgia -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: text/enriched Size: 1326 bytes Desc: not available URL: From jose.martin at wanadoo.es Fri Jan 30 15:22:58 2004 From: jose.martin at wanadoo.es (=?iso-8859-1?Q?Jos=E9_A_Mart=EDn_H?=) Date: Fri, 30 Jan 2004 21:22:58 +0100 Subject: [SciPy-user] HELP !!! sci py dost not work !!! References: <002401c3e76c$e46ae200$52806750@CASA> Message-ID: <004001c3e76e$dad38560$52806750@CASA> Can i build scipy in windows Xp MSVC ? it is easy to do this ? thanks... ----- Original Message ----- From: Christopher Fonnesbeck To: SciPy Users List Sent: Friday, January 30, 2004 9:05 PM Subject: Re: [SciPy-user] HELP !!! sci py dost not work !!! On Jan 30, 2004, at 3:08 PM, Jos? A Mart?n H wrote: IDLE 1.0.2 >>> import scipy Traceback (most recent call last): File "", line 1, in -toplevel- import scipy File "C:\Python23\Lib\site-packages\scipy\__init__.py", line 11, in -toplevel- from scipy_base import * ImportError: No module named scipy_base >>> i have upgraded to python 2.3 and downloaded the last binaries for scipy !!!. You are much better off compiling scipy from CVS than using the binaries. C. -- Christopher J. Fonnesbeck ( c h r i s @ f o n n e s b e c k . o r g ) Georgia Cooperative Fish & Wildlife Research Unit, University of Georgia ------------------------------------------------------------------------------ _______________________________________________ SciPy-user mailing list SciPy-user at scipy.net http://www.scipy.net/mailman/listinfo/scipy-user -------------- next part -------------- An HTML attachment was scrubbed... URL: From chris at fonnesbeck.org Fri Jan 30 15:17:00 2004 From: chris at fonnesbeck.org (Christopher Fonnesbeck) Date: Fri, 30 Jan 2004 15:17:00 -0500 Subject: [SciPy-user] HELP !!! sci py dost not work !!! In-Reply-To: <004001c3e76e$dad38560$52806750@CASA> References: <002401c3e76c$e46ae200$52806750@CASA> <004001c3e76e$dad38560$52806750@CASA> Message-ID: <42CA3D2A-5361-11D8-94E2-000A956FDAC0@fonnesbeck.org> On Jan 30, 2004, at 3:22 PM, Jos? A Mart?n H wrote: > Can i build scipy in windows Xp? MSVC ? > ? > it is easy to do this ? > ? > thanks... Well, its not that hard ... I dont know about "easy". Just make sure you have the prerequisites, and grab the CVS files, following these instructions: http://www.scipy.org/site_content/tutorials/CVSInstruct There is an INSTALL.txt file in the scipy directory when the download completes. C. -- Christopher J. Fonnesbeck ( c h r i s @ f o n n e s b e c k . o r g ) Georgia Cooperative Fish & Wildlife Research Unit, University of Georgia -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: text/enriched Size: 873 bytes Desc: not available URL: From joseamartin at cantv.net Fri Jan 30 14:54:42 2004 From: joseamartin at cantv.net (=?iso-8859-1?Q?Jos=E9_A_Mart=EDn_H?=) Date: Fri, 30 Jan 2004 20:54:42 +0100 Subject: [SciPy-user] Help, python 2.3.3 scipy PILUTIL Problem Message-ID: <000c01c3e76a$e7dc4cf0$52806750@CASA> Hi, i am user of scipy and PIL, scipy includes a module called PILUTIL for working wiht pil I have upgraded to python 2.3 and scipy for python 2.3 but my programs dosnt work i get an error when using PILUTIL.py , i dont have the ( scipy_base ) package and pilutil import some things from it. I can i do ? i really need the module pilutil thanks.... -------------- next part -------------- An HTML attachment was scrubbed... URL: From joseamartin at cantv.net Fri Jan 30 15:08:03 2004 From: joseamartin at cantv.net (=?iso-8859-1?Q?Jos=E9_A_Mart=EDn_H?=) Date: Fri, 30 Jan 2004 21:08:03 +0100 Subject: [SciPy-user] H E L P !!! PLEASE Message-ID: <000f01c3e76c$c560de50$52806750@CASA> IDLE 1.0.2 >>> import scipy Traceback (most recent call last): File "", line 1, in -toplevel- import scipy File "C:\Python23\Lib\site-packages\scipy\__init__.py", line 11, in -toplevel- from scipy_base import * ImportError: No module named scipy_base >>> i have upgraded to python 2.3 and downloaded the last binaries for scipy !!!. -------------- next part -------------- An HTML attachment was scrubbed... URL: