From chris at fonnesbeck.org Mon Nov 3 17:34:39 2003 From: chris at fonnesbeck.org (Christopher Fonnesbeck) Date: Mon, 3 Nov 2003 17:34:39 -0500 Subject: [SciPy-user] build error related to fblas.so on OSX Message-ID: I am encountering a gcc error while trying to build either the latest release or the CVS of scipy: gcc -Wl,-F. -Wl,-F. -bundle -framework Python build/temp.darwin-7.0.0-Power_Macintosh-2.3/System/Library/Frameworks/ Python.framework/Versions/2.3/lib/python2.3/site-packages/f2py2e/src/ fortranobject.o build/temp.darwin-7.0.0-Power_Macintosh-2.3/build/temp.darwin-7.0.0- Power_Macintosh-2.3/fblasmodule.o -L/usr/local/lib -Lbuild/temp.darwin-7.0.0-Power_Macintosh-2.3 -Lbuild/temp.darwin-7.0.0-Power_Macintosh-2.3 -L/usr/local/lib/gcc/powerpc-apple-darwin6.6/3.4 -L/usr/local/lib -lfblas_f2py -llapack -lf77blas -lcblas -latlas -lc_misc -lcephes -lrootfind -lg2c -o build/lib.darwin-7.0.0-Power_Macintosh-2.3/scipy/linalg/fblas.so ld: Undefined symbols: restFP saveFP restVEC_vr10 saveVEC_vr11 error: command 'gcc' failed with exit status 1 I am on OSX, using python2.3 and using the blas libraries that are bundled with the OSX developer tools. Any insight as to how I might get around these errors are most welcome. 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 bob at redivi.com Mon Nov 3 17:44:04 2003 From: bob at redivi.com (Bob Ippolito) Date: Mon, 3 Nov 2003 17:44:04 -0500 Subject: [SciPy-user] build error related to fblas.so on OSX In-Reply-To: References: Message-ID: <39F8E962-0E4F-11D8-BBD5-000A95686CD8@redivi.com> On Nov 3, 2003, at 5:34 PM, Christopher Fonnesbeck wrote: > I am encountering a gcc error while trying to build either the latest > release or the CVS of scipy: > > gcc -Wl,-F. -Wl,-F. -bundle -framework Python > build/temp.darwin-7.0.0-Power_Macintosh-2.3/System/Library/Frameworks/ > Python.framework/Versions/2.3/lib/python2.3/site-packages/f2py2e/src/ > fortranobject.o > build/temp.darwin-7.0.0-Power_Macintosh-2.3/build/temp.darwin-7.0.0- > Power_Macintosh-2.3/fblasmodule.o -L/usr/local/lib > -Lbuild/temp.darwin-7.0.0-Power_Macintosh-2.3 > -Lbuild/temp.darwin-7.0.0-Power_Macintosh-2.3 > -L/usr/local/lib/gcc/powerpc-apple-darwin6.6/3.4 -L/usr/local/lib > -lfblas_f2py -llapack -lf77blas -lcblas -latlas -lc_misc -lcephes > -lrootfind -lg2c -o > build/lib.darwin-7.0.0-Power_Macintosh-2.3/scipy/linalg/fblas.so > ld: Undefined symbols: > restFP > saveFP > restVEC_vr10 > saveVEC_vr11 > error: command 'gcc' failed with exit status 1 > > I am on OSX, using python2.3 and using the blas libraries that are > bundled with the OSX developer tools. Any insight as to how I might > get around these errors are most welcome. I believe you need to add the -faltivec compiler flag. -bob From chris at fonnesbeck.org Mon Nov 3 18:58:00 2003 From: chris at fonnesbeck.org (Christopher Fonnesbeck) Date: Mon, 3 Nov 2003 18:58:00 -0500 Subject: [SciPy-user] build error related to fblas.so on OSX Message-ID: <8DFC3D1E-0E59-11D8-9C90-000A956FDAC0@fonnesbeck.org> Great, thanks! Can I specify this flag on the command line when I call python setup install? 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 bob at redivi.com Mon Nov 3 19:04:30 2003 From: bob at redivi.com (Bob Ippolito) Date: Mon, 3 Nov 2003 19:04:30 -0500 Subject: [SciPy-user] build error related to fblas.so on OSX In-Reply-To: <8DFC3D1E-0E59-11D8-9C90-000A956FDAC0@fonnesbeck.org> References: <8DFC3D1E-0E59-11D8-9C90-000A956FDAC0@fonnesbeck.org> Message-ID: <766A1FA0-0E5A-11D8-BBD5-000A95686CD8@redivi.com> On Nov 3, 2003, at 6:58 PM, Christopher Fonnesbeck wrote: > Great, thanks! Can I specify this flag on the command line when I > call python setup install? Um.. it should probably go in the extra_compile_args argument of the Extension(...) in the setup.py. But maybe if you don't want to do that, then "setenv CFLAGS -faltivec" (this is tcsh syntax) before compiling that may do the trick. -bob From chris at fonnesbeck.org Mon Nov 3 19:31:35 2003 From: chris at fonnesbeck.org (Christopher Fonnesbeck) Date: Mon, 3 Nov 2003 19:31:35 -0500 Subject: [SciPy-user] build error related to fblas.so on OSX Message-ID: <3F242FD8-0E5E-11D8-9C90-000A956FDAC0@fonnesbeck.org> Unfortunately adding the -faltivec argument did not prevent the error: gcc -Wl,-F. -Wl,-F. -bundle -framework Python -faltivec build/temp.darwin-7.0.0-Power_Macintosh-2.3/System/Library/Frameworks/ Python.framework/Versions/2.3/lib/python2.3/site-packages/f2py2e/src/ fortranobject.o build/temp.darwin-7.0.0-Power_Macintosh-2.3/build/temp.darwin-7.0.0- Power_Macintosh-2.3/fblasmodule.o -L/usr/local/lib -Lbuild/temp.darwin-7.0.0-Power_Macintosh-2.3 -Lbuild/temp.darwin-7.0.0-Power_Macintosh-2.3 -L/usr/local/lib/gcc/powerpc-apple-darwin6.6/3.4 -L/usr/local/lib -lfblas_f2py -llapack -lf77blas -lcblas -latlas -lc_misc -lcephes -lrootfind -lg2c -o build/lib.darwin-7.0.0-Power_Macintosh-2.3/scipy/linalg/fblas.so ld: Undefined symbols: restFP saveFP restVEC_vr10 saveVEC_vr11 error: command 'gcc' failed with exit status 1 -- 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 serquiciamailing at fastimap.com Tue Nov 4 12:45:41 2003 From: serquiciamailing at fastimap.com (Santiago Erquicia) Date: Tue, 04 Nov 2003 11:45:41 -0600 Subject: [SciPy-user] Problem with arange Message-ID: <3FA7E5C5.6040200@fastimap.com> I'm using Numeric.arange and I have different results depending on the numbers I'm using. For example: arange(9, 12.1, 0.1, Float) array([ 9. , 9.1, 9.2, 9.3, 9.4, 9.5, 9.6, 9.7, 9.8, 9.9, 10. , 10.1, 10.2, 10.3, 10.4, 10.5, 10.6, 10.7, 10.8, 10.9, 11. , 11.1, 11.2, 11.3, 11.4, 11.5, 11.6, 11.7, 11.8, 11.9, 12. ]) arange(18, 21.1, 0.1, Float) array([ 18. , 18.1, 18.2, 18.3, 18.4, 18.5, 18.6, 18.7, 18.8, 18.9, 19. , 19.1, 19.2, 19.3, 19.4, 19.5, 19.6, 19.7, 19.8, 19.9, 20. , 20.1, 20.2, 20.3, 20.4, 20.5, 20.6, 20.7, 20.8, 20.9, 21. , 21.1]) The difference is that the first one doesn't include the max value, but the second one do this. Do you know how to solve this? I'm using python enthought edition. Thanks in advance From kern at ugcs.caltech.edu Tue Nov 4 13:03:24 2003 From: kern at ugcs.caltech.edu (Robert Kern) Date: Tue, 4 Nov 2003 10:03:24 -0800 Subject: [SciPy-user] Problem with arange In-Reply-To: <3FA7E5C5.6040200@fastimap.com> References: <3FA7E5C5.6040200@fastimap.com> Message-ID: <20031104180324.GA4499@dinar.ugcs.caltech.edu> On Tue, Nov 04, 2003 at 11:45:41AM -0600, Santiago Erquicia wrote: > I'm using Numeric.arange and I have different results depending on the > numbers I'm using. > > For example: > arange(9, 12.1, 0.1, Float) > array([ 9. , 9.1, 9.2, 9.3, 9.4, 9.5, 9.6, 9.7, 9.8, > 9.9, 10. , > 10.1, 10.2, 10.3, 10.4, 10.5, 10.6, 10.7, 10.8, > 10.9, 11. , > 11.1, 11.2, 11.3, 11.4, 11.5, 11.6, 11.7, 11.8, > 11.9, 12. ]) > > > arange(18, 21.1, 0.1, Float) > array([ 18. , 18.1, 18.2, 18.3, 18.4, 18.5, 18.6, 18.7, 18.8, > 18.9, 19. , > 19.1, 19.2, 19.3, 19.4, 19.5, 19.6, 19.7, 19.8, > 19.9, 20. , > 20.1, 20.2, 20.3, 20.4, 20.5, 20.6, 20.7, 20.8, > 20.9, 21. , > 21.1]) > > The difference is that the first one doesn't include the max value, but > the second one do this. > > Do you know how to solve this? Either use integral steps and divide by 10, or use 12.05 and 21.05 as your maximum values. Floating point weirds numbers. > I 'm using python enthought edition. > > Thanks in advance -- Robert Kern kern at ugcs.caltech.edu "In the fields of Hell where the grass grows high Are the graves of dreams allowed to die." -- Richard Harter From falted at openlc.org Tue Nov 4 13:16:55 2003 From: falted at openlc.org (Francesc Alted) Date: Tue, 4 Nov 2003 19:16:55 +0100 Subject: [SciPy-user] Problem with arange In-Reply-To: <3FA7E5C5.6040200@fastimap.com> References: <3FA7E5C5.6040200@fastimap.com> Message-ID: <200311041916.55452.falted@openlc.org> A Dimarts 04 Novembre 2003 18:45, Santiago Erquicia va escriure: > I'm using Numeric.arange and I have different results depending on the > numbers I'm using. > > For example: > arange(9, 12.1, 0.1, Float) > arange(18, 21.1, 0.1, Float) > > The difference is that the first one doesn't include the max value, but > the second one do this. This is a typical problem with float representation on a machine. Look at that: >>> 12.1 12.1 >>> 21.1 21.100000000000001 so, many floating point numbers has not an exact representation. In your case, 21 < 21.100000000000001, so these rounding problems are causing that kind of problems. A workaround on that is to define a very small number compared with your step size (but still larger than the machine precision for this type!), and this will get solved: >>> epsilon = 1e-15 >>> arange(9, 12.1+epsilon, 0.1, Float) array([ 9. , 9.1, 9.2, 9.3, 9.4, 9.5, 9.6, 9.7, 9.8, 9.9, 10. , 10.1, 10.2, 10.3, 10.4, 10.5, 10.6, 10.7, 10.8, 10.9, 11. , 11.1, 11.2, 11.3, 11.4, 11.5, 11.6, 11.7, 11.8, 11.9, 12. , 12.1]) >>> arange(18, 21.1+epsilon, 0.1, Float) array([ 18. , 18.1, 18.2, 18.3, 18.4, 18.5, 18.6, 18.7, 18.8, 18.9, 19. , 19.1, 19.2, 19.3, 19.4, 19.5, 19.6, 19.7, 19.8, 19.9, 20. , 20.1, 20.2, 20.3, 20.4, 20.5, 20.6, 20.7, 20.8, 20.9, 21. , 21.1]) (or do a substraction if you don't want the upper limit to be included, in the pure python tradition) Cheers, -- Francesc Alted From lanceboyle at myrealbox.com Wed Nov 5 23:39:58 2003 From: lanceboyle at myrealbox.com (Lance Boyle) Date: Wed, 5 Nov 2003 21:39:58 -0700 Subject: [SciPy-user] Problem with arange In-Reply-To: <3FA7E5C5.6040200@fastimap.com> Message-ID: <46F59BAE-1013-11D8-A79A-003065F93FF0@myrealbox.com> On Tuesday, Nov 4, 2003, at 10:45 America/Phoenix, Santiago Erquicia wrote: > I'm using Numeric.arange and I have different results depending on the > numbers I'm using. > > For example: > arange(9, 12.1, 0.1, Float) > array([ 9. , 9.1, 9.2, 9.3, 9.4, 9.5, 9.6, 9.7, 9.8, > 9.9, 10. , > 10.1, 10.2, 10.3, 10.4, 10.5, 10.6, 10.7, 10.8, > 10.9, 11. , > 11.1, 11.2, 11.3, 11.4, 11.5, 11.6, 11.7, 11.8, > 11.9, 12. ]) > > > arange(18, 21.1, 0.1, Float) > array([ 18. , 18.1, 18.2, 18.3, 18.4, 18.5, 18.6, 18.7, 18.8, > 18.9, 19. , > 19.1, 19.2, 19.3, 19.4, 19.5, 19.6, 19.7, 19.8, > 19.9, 20. , > 20.1, 20.2, 20.3, 20.4, 20.5, 20.6, 20.7, 20.8, > 20.9, 21. , > 21.1]) > > The difference is that the first one doesn't include the max value, > but the second one do this. > > Do you know how to solve this? > > I'm using python enthought edition. It's everywhere: >>> arange(1.0e-18, 2.0e-18, 1.0e-19, Float) array([ 1.00000000e-18, 1.10000000e-18, 1.20000000e-18, 1.30000000e-18, 1.40000000e-18, 1.50000000e-18, 1.60000000e-18, 1.70000000e-18, 1.80000000e-18, 1.90000000e-18, 2.00000000e-18]) >>> arange(2.0e-18, 3.0e-18, 1.0e-19, Float) array([ 2.00000000e-18, 2.10000000e-18, 2.20000000e-18, 2.30000000e-18, 2.40000000e-18, 2.50000000e-18, 2.60000000e-18, 2.70000000e-18, 2.80000000e-18, 2.90000000e-18]) This kind of floating point issue is present in some form on any platform and language. However, Numeric.arange() makes it possible to "automate" the error. A language with ambitions to be a numerical computing platform should not make it this easy. Numeric.arange() is a dangerous function and should be eliminated. Bottom line: Don't use it unless you know without a doubt that your "epsilon" will always work. Jerry From lanceboyle at myrealbox.com Wed Nov 5 23:56:48 2003 From: lanceboyle at myrealbox.com (Lance Boyle) Date: Wed, 5 Nov 2003 21:56:48 -0700 Subject: [SciPy-user] Problem with arange In-Reply-To: <3FA7E5C5.6040200@fastimap.com> Message-ID: On Tuesday, Nov 4, 2003, at 10:45 America/Phoenix, Santiago Erquicia wrote: > I'm using Numeric.arange and I have different results depending on the > numbers I'm using. > > For example: > arange(9, 12.1, 0.1, Float) > array([ 9. , 9.1, 9.2, 9.3, 9.4, 9.5, 9.6, 9.7, 9.8, > 9.9, 10. , > 10.1, 10.2, 10.3, 10.4, 10.5, 10.6, 10.7, 10.8, > 10.9, 11. , > 11.1, 11.2, 11.3, 11.4, 11.5, 11.6, 11.7, 11.8, > 11.9, 12. ]) > > > arange(18, 21.1, 0.1, Float) > array([ 18. , 18.1, 18.2, 18.3, 18.4, 18.5, 18.6, 18.7, 18.8, > 18.9, 19. , > 19.1, 19.2, 19.3, 19.4, 19.5, 19.6, 19.7, 19.8, > 19.9, 20. , > 20.1, 20.2, 20.3, 20.4, 20.5, 20.6, 20.7, 20.8, > 20.9, 21. , > 21.1]) > > The difference is that the first one doesn't include the max value, > but the second one do this. > > Do you know how to solve this? > > I'm using python enthought edition. > For your further amusement: >>> 3.1 3.1000000000000001 >>> 4.1 4.0999999999999996 >>> 31.0 * 0.1 3.1000000000000001 >>> 41.0 * 0.1 4.1000000000000005 >>> arange(1, 3.1, 0.1, Float) array([ 1. , 1.1, 1.2, 1.3, 1.4, 1.5, 1.6, 1.7, 1.8, 1.9, 2. , 2.1, 2.2, 2.3, 2.4, 2.5, 2.6, 2.7, 2.8, 2.9, 3. ]) >>> arange(3.0, 3.1, 0.1, Float) array([ 3. , 3.1]) There appears to be implementation issues with arange(). FWIW, whole numbers up to some large number in magnitude (..., -2.0, -1.0, 0.0, 1.0, 2.0, ...) are supposed to be represented exactly in IEEE arithmetic. Jerry From fperez at colorado.edu Thu Nov 6 01:49:51 2003 From: fperez at colorado.edu (Fernando Perez) Date: Wed, 05 Nov 2003 23:49:51 -0700 Subject: [SciPy-user] Problem with arange In-Reply-To: <46F59BAE-1013-11D8-A79A-003065F93FF0@myrealbox.com> References: <46F59BAE-1013-11D8-A79A-003065F93FF0@myrealbox.com> Message-ID: <3FA9EF0F.4080308@colorado.edu> Lance Boyle wrote: > This kind of floating point issue is present in some form on any > platform and language. However, Numeric.arange() makes it possible to > "automate" the error. A language with ambitions to be a > numerical computing platform should not make it this easy. > Numeric.arange() is a dangerous function and should be eliminated. > > Bottom line: Don't use it unless you know without a doubt that your > "epsilon" will always work. Flamebait aside, here's the implementation I use for my own work, which is part of ipython. After this thread I noticed my original frange() had the same problem as arange(), but because of the way it's written, it was trivial to fix. I guess I prefer to fix a useful function which has a problem rather than eliminating every piece of code which isn't perfect. Feel free to use it if you like, or if you already use ipython, it's in IPython.numutils. Note that my frange defaults to CLOSED intervals, because that's how I tend to use it most (in fact, I use the npts=N option ~100% of the time). For plotting, I typically want the endpoints included. But you can set the default to open trivially by modifying one line below. def frange(xini,xfin=None,delta=None,**kw): """frange([start,] stop[, step, keywords]) -> list of floats Return a Numeric array() containing a progression of floats. Similar to arange(), but defaults to a closed interval. frange(x0, x1) returns [x0, x0+1, x0+2, ..., x1]; start defaults to 0, and the endpoint *is included*. This behavior is different from that of range() and arange(). This is deliberate, since frange will probably be more useful for generating lists of points for function evaluation, and endpoints are often desired in this use. The usual behavior of range() can be obtained by setting the keyword 'closed=0', in this case frange() basically becomes arange(). When step is given, it specifies the increment (or decrement). All arguments can be floating point numbers. frange(x0,x1,d) returns [x0,x0+d,x0+2d,...,xfin] where xfin<=x1. frange can also be called with the keyword 'npts'. This sets the number of points the list should contain (and overrides the value 'step' might have been given). arange() doesn't offer this option. Examples: >>> frange(3) array([ 0., 1., 2., 3.]) >>> frange(3,closed=0) array([ 0., 1., 2.]) >>> frange(1,6,2) array([1, 3, 5]) >>> frange(1,6.5,npts=5) array([ 1. , 2.375, 3.75 , 5.125, 6.5 ]) """ #defaults kw.setdefault('closed',1) endpoint = kw['closed'] != 0 # funny logic to allow the *first* argument to be optional (like range()) # This was modified with a simpler version from a similar frange() found # at http://aspn.activestate.com/ASPN/Cookbook/Python/Recipe/66472 if xfin == None: xfin = xini + 0.0 xini = 0.0 if delta == None: delta = 1.0 # compute # of points, spacing and return final list try: npts=kw['npts'] delta=(xfin-xini)/float(npts-endpoint) except KeyError: # round() gets npts right even with the vagaries of floating point. npts=int(round((xfin-xini)/delta+endpoint)) return arange(npts)*delta+xini Cheers, f. From falted at openlc.org Thu Nov 6 03:53:38 2003 From: falted at openlc.org (Francesc Alted) Date: Thu, 6 Nov 2003 09:53:38 +0100 Subject: [SciPy-user] Problem with arange In-Reply-To: <3FA9EF0F.4080308@colorado.edu> References: <46F59BAE-1013-11D8-A79A-003065F93FF0@myrealbox.com> <3FA9EF0F.4080308@colorado.edu> Message-ID: <200311060953.39180.falted@openlc.org> A Dijous 06 Novembre 2003 07:49, Fernando Perez va escriure: > def frange(xini,xfin=None,delta=None,**kw): > """frange([start,] stop[, step, keywords]) -> list of floats > > Return a Numeric array() containing a progression of floats. Similar > to arange(), but defaults to a closed interval. Uau, that's a very nice piece of code: simple and effective! What about modifying arange in numarray/Numeric in that way? I think arange with floating parameters is so common that such an addition would save many headaches to users. Cheers, -- Francesc Alted From wagner.nils at vdi.de Thu Nov 6 04:54:09 2003 From: wagner.nils at vdi.de (Nils Wagner) Date: Thu, 06 Nov 2003 10:54:09 +0100 Subject: [SciPy-user] io.savemat Message-ID: <20031106095801.2BF103EB20@www.scipy.com> Hi all, How can I store the matrix >>> A array([[ 0.94810236, 0.81994671, 0.90014333], [ 0.79542607, 0.89239591, 0.37511933], [ 0.73988128, 0.19856018, 0.52253771]]) with io.savemat ? A small example would be appreciated. BTW, Is the output restricted to matlab Version 4 ? Nils From oliphant at ee.byu.edu Thu Nov 6 05:34:41 2003 From: oliphant at ee.byu.edu (Travis Oliphant) Date: Thu, 06 Nov 2003 03:34:41 -0700 Subject: [SciPy-user] Problem with arange In-Reply-To: <200311060953.39180.falted@openlc.org> References: <46F59BAE-1013-11D8-A79A-003065F93FF0@myrealbox.com> <3FA9EF0F.4080308@colorado.edu> <200311060953.39180.falted@openlc.org> Message-ID: <3FAA23C1.7060308@ee.byu.edu> Francesc Alted wrote: > A Dijous 06 Novembre 2003 07:49, Fernando Perez va escriure: > >>def frange(xini,xfin=None,delta=None,**kw): >> """frange([start,] stop[, step, keywords]) -> list of floats >> >> Return a Numeric array() containing a progression of floats. Similar >>to arange(), but defaults to a closed interval. > > Apparently most are unfamiliar with SciPy's linspace and r_[] functionality (I never use arange) a = r[-1:1:100j] gives 100 points between -1 and 1 a = linspace(-1,1,100) # endpoint defaults to 1 -Travis O. From fperez at colorado.edu Thu Nov 6 13:53:02 2003 From: fperez at colorado.edu (Fernando Perez) Date: Thu, 06 Nov 2003 11:53:02 -0700 Subject: [SciPy-user] Problem with arange In-Reply-To: <3FAA23C1.7060308@ee.byu.edu> References: <46F59BAE-1013-11D8-A79A-003065F93FF0@myrealbox.com> <3FA9EF0F.4080308@colorado.edu> <200311060953.39180.falted@openlc.org> <3FAA23C1.7060308@ee.byu.edu> Message-ID: <3FAA988E.5080708@colorado.edu> Travis Oliphant wrote: > > Francesc Alted wrote: > >>A Dijous 06 Novembre 2003 07:49, Fernando Perez va escriure: >> >> >>>def frange(xini,xfin=None,delta=None,**kw): >>> """frange([start,] stop[, step, keywords]) -> list of floats >>> >>> Return a Numeric array() containing a progression of floats. Similar >>>to arange(), but defaults to a closed interval. >> >> > > Apparently most are unfamiliar with SciPy's linspace and r_[] > functionality (I never use arange) > > a = r[-1:1:100j] gives 100 points between -1 and 1 > a = linspace(-1,1,100) # endpoint defaults to 1 Well, I actually wrote this code long before I was aware even of scipy :) But you're right, I didn't know about those. Still, my little frange is useful when I don't want to load all of scipy but just need Numeric + a few simple utilities for plotting or testing things. Thanks for the pointer, though. As time goes by, I'm converting more and more of my daily use to scipy... Best, f From fperez at colorado.edu Thu Nov 6 13:54:39 2003 From: fperez at colorado.edu (Fernando Perez) Date: Thu, 06 Nov 2003 11:54:39 -0700 Subject: [SciPy-user] Problem with arange In-Reply-To: <200311060953.39180.falted@openlc.org> References: <46F59BAE-1013-11D8-A79A-003065F93FF0@myrealbox.com> <3FA9EF0F.4080308@colorado.edu> <200311060953.39180.falted@openlc.org> Message-ID: <3FAA98EF.4010100@colorado.edu> Francesc Alted wrote: > A Dijous 06 Novembre 2003 07:49, Fernando Perez va escriure: > >>def frange(xini,xfin=None,delta=None,**kw): >> """frange([start,] stop[, step, keywords]) -> list of floats >> >> Return a Numeric array() containing a progression of floats. Similar >>to arange(), but defaults to a closed interval. > > > Uau, that's a very nice piece of code: simple and effective! What about > modifying arange in numarray/Numeric in that way? I think arange with > floating parameters is so common that such an addition would save many > headaches to users. FWIW, anyone who wants it is free to use that code (as is or modified) wherever they want, including incorporating it into numarray. Cheers, f. ps. Thanks for the nice comment :) From wagner.nils at vdi.de Fri Nov 7 03:25:27 2003 From: wagner.nils at vdi.de (Nils Wagner) Date: Fri, 07 Nov 2003 09:25:27 +0100 Subject: [SciPy-user] Usage of savemat Message-ID: <20031107082924.E99E13EB20@www.scipy.com> Hi all, I am sorry, but I still have some trouble with savemat. Please reply. 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. >>> from scipy import * >>> from RandomArray import * >>> A = rand(3,3) >>> B ={1:A} >>> B[1] array([[ 0.92210221, 0.72158247, 0.23778352], [ 0.4173221 , 0.05069149, 0.26121452], [ 0.02657581, 0.17013131, 0.2672503 ]]) >>> io.savemat('randmat',B) Traceback (most recent call last): File "", line 1, in ? File "/usr/lib/python2.3/site-packages/scipy/io/mio.py", line 818, in savemat fid.fwrite([var.shape[1], var.shape[0], imagf, len(variable)+1],'int') TypeError: len() of unsized object >>> io.savemat('randmat',B[1]) Traceback (most recent call last): File "", line 1, in ? File "/usr/lib/python2.3/site-packages/scipy/io/mio.py", line 797, in savemat for variable in dict.keys(): AttributeError: keys >>> Thanks in advance. Nils From pearu at scipy.org Fri Nov 7 03:33:52 2003 From: pearu at scipy.org (Pearu Peterson) Date: Fri, 7 Nov 2003 02:33:52 -0600 (CST) Subject: [SciPy-user] Usage of savemat In-Reply-To: <20031107082924.E99E13EB20@www.scipy.com> Message-ID: On Fri, 7 Nov 2003, Nils Wagner wrote: > Hi all, > > I am sorry, but I still have some trouble with savemat. > Please reply. > > 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. > >>> from scipy import * > >>> from RandomArray import * > >>> A = rand(3,3) > >>> B ={1:A} ^^^ see io.savemat.__doc__ So, try >>> B ={'A':A} ... Pearu From ariciputi at pito.com Fri Nov 7 04:31:52 2003 From: ariciputi at pito.com (Andrea Riciputi) Date: Fri, 7 Nov 2003 10:31:52 +0100 Subject: [SciPy-user] Help on client-server mode. (OT?) Message-ID: <38730CD4-1105-11D8-9181-000393933E4E@pito.com> Hi there, I need a little help here and even if it's not completely related to the list I think you can help me. I'm working on a number crunching application partly written in C, partly in Python(+Scipy). The application is becoming more and more time consuming and soon it'll takes forever to run on my laptop. So I was wondering if it was possible to run time-consuming tasks (namely the C-side of my code) on a remote host (a workstation I mean) keeping all the Python code (that I use mostly as a control language) on my laptop. In this picture the Python code has to work in a sort of client-server mode: tell the server to start-up some CPU-intensive code (previously compiled and linked by means of SWIG), wait for the answer, receive the returning results (namely one or more Scipy matrix). Is it possible? And how? Thanks in advance, Andrea. --- Andrea Riciputi "Science is like sex: sometimes something useful comes out, but that is not the reason we are doing it" -- (Richard Feynman) From jdhunter at ace.bsd.uchicago.edu Fri Nov 7 08:19:02 2003 From: jdhunter at ace.bsd.uchicago.edu (John Hunter) Date: Fri, 07 Nov 2003 07:19:02 -0600 Subject: [SciPy-user] Help on client-server mode. (OT?) In-Reply-To: <38730CD4-1105-11D8-9181-000393933E4E@pito.com> (Andrea Riciputi's message of "Fri, 7 Nov 2003 10:31:52 +0100") References: <38730CD4-1105-11D8-9181-000393933E4E@pito.com> Message-ID: >>>>> "Andrea" == Andrea Riciputi writes: Andrea> So I was wondering if it was possible to run Andrea> time-consuming tasks (namely the C-side of my code) on a Andrea> remote host (a workstation I mean) keeping all the Python Andrea> code (that I use mostly as a control language) on my Andrea> laptop. In this picture the Python code has to work in a Andrea> sort of client-server mode: tell the server to start-up Andrea> some CPU-intensive code (previously compiled and linked by Andrea> means of SWIG), wait for the answer, receive the returning Andrea> results (namely one or more Scipy matrix). Andrea> Is it possible? And how? You may want to look at Pyro which allows remote communication of objects. It's a very nice package. http://pyro.sourceforge.net Here's a simple example. 'mother' is the hostname of the remote number cruncher and I'm using port 2358 to communicate Server side import sys from Pyro.ext import remote_nons class RemoteCruncher: def get_shape(self, a): return a.shape remote_nons.provide_server_object( RemoteCruncher(), 'RemoteCruncher', 'mother', 2358) print 'Waiting for requests.' sys.exit(remote_nons.handle_requests()) Client side from Pyro.ext import remote_nons from Numeric import array o = remote_nons.get_server_object('RemoteCruncher', 'mother', 2358) a = array([1,2,3]) print o.get_shape(a) Hope this helps! JDH From ariciputi at pito.com Fri Nov 7 10:35:03 2003 From: ariciputi at pito.com (Andrea Riciputi) Date: Fri, 7 Nov 2003 16:35:03 +0100 Subject: [SciPy-user] Help on client-server mode. (OT?) In-Reply-To: Message-ID: It seems interesting, I'll give it a try. But as far as I can understand I must have Python installed on both side (server and client) in order to get it to work. Is it possible to get the same result (obviously not using Pyro) without Python on the server side?? (Well... I think not.) Thanks, Andrea. On Friday, Nov 7, 2003, at 14:19 Europe/Rome, John Hunter wrote: > You may want to look at Pyro which allows remote communication of > objects. It's a very nice package. > > http://pyro.sourceforge.net > > Here's a simple example. 'mother' is the hostname of the remote > number cruncher and I'm using port 2358 to communicate > > Server side > > import sys > from Pyro.ext import remote_nons > > class RemoteCruncher: > def get_shape(self, a): > return a.shape > > remote_nons.provide_server_object( > RemoteCruncher(), 'RemoteCruncher', 'mother', 2358) > > print 'Waiting for requests.' > sys.exit(remote_nons.handle_requests()) > > Client side > > from Pyro.ext import remote_nons > from Numeric import array > > o = remote_nons.get_server_object('RemoteCruncher', 'mother', 2358) > > a = array([1,2,3]) > print o.get_shape(a) > > Hope this helps! > > JDH > _______________________________________________ > SciPy-user mailing list > SciPy-user at scipy.net > http://www.scipy.net/mailman/listinfo/scipy-user > > --- Andrea Riciputi "Science is like sex: sometimes something useful comes out, but that is not the reason we are doing it" -- (Richard Feynman) --- Andrea Riciputi "Science is like sex: sometimes something useful comes out, but that is not the reason we are doing it" -- (Richard Feynman) From jdhunter at ace.bsd.uchicago.edu Fri Nov 7 10:39:21 2003 From: jdhunter at ace.bsd.uchicago.edu (John Hunter) Date: Fri, 07 Nov 2003 09:39:21 -0600 Subject: [SciPy-user] Help on client-server mode. (OT?) In-Reply-To: (Andrea Riciputi's message of "Fri, 7 Nov 2003 16:35:03 +0100") References: Message-ID: >>>>> "Andrea" == Andrea Riciputi writes: Andrea> It seems interesting, I'll give it a try. But as far as I Andrea> can understand I must have Python installed on both side Andrea> (server and client) in order to get it to work. Is it Andrea> possible to get the same result (obviously not using Pyro) Andrea> without Python on the server side?? (Well... I think Andrea> not.) Sure, you'll need to have your program listening on some socket for some preformatted data coming in, and use the python socket module to talk to it. This is harder than using a solution that has already been implemented but certainly possible. If it were me, though, I would write a python wrapper around my C code (SWIG or boost:::python for C++) and let pyro do the networking for me. It shouldn't be hard to install python on the server side. This introduces some overhead and so will likely be slower than a hand-rolled socket communication directly to your C routines. But probably easier to get up and running. Also, if most of the time is spent crunching numbers rather than passing data over the network, the extra overhead of the server side pyro interface will be small. JDH From falted at openlc.org Fri Nov 7 11:32:07 2003 From: falted at openlc.org (Francesc Alted) Date: Fri, 7 Nov 2003 17:32:07 +0100 Subject: [SciPy-user] Help on client-server mode. (OT?) In-Reply-To: References: Message-ID: <200311071732.07556.falted@openlc.org> A Divendres 07 Novembre 2003 16:35, Andrea Riciputi va escriure: > It seems interesting, I'll give it a try. But as far as I can > understand I must have Python installed on both side (server and > client) in order to get it to work. Is it possible to get the same > result (obviously not using Pyro) without Python on the server side?? > (Well... I think not.) There are a lot of free solutions out there like CORBA (for example omniORB), XML-RPC or even Jabber allows one to create the server in C(++) and ask for the results in any language that supports these protocols (as, for example, python :-). If you have little need for efficient communications, XML-RPC or, perhaps, Jabber (which is ultimately based on XML too) are very simple to use. However, if you do need an *efficient* protocol, CORBA is a far better approach (at the cost of increased complexity). One alternative which represents a good balance between efficency and simplicity would be spread (www.spread.org). Cheers, -- Francesc Alted From chris at fonnesbeck.org Tue Nov 11 08:06:18 2003 From: chris at fonnesbeck.org (Christopher Fonnesbeck) Date: Tue, 11 Nov 2003 08:06:18 -0500 Subject: [SciPy-user] compile failure on OSX Message-ID: I now have wxPython working on OSX 10.3, but the scipy build still fails. Here is the error: gcc -Wl,-F. -Wl,-F. -bundle -framework Python -faltivec build/temp.darwin-7.0.0-Power_Macintosh-2.3/System/Library/Frameworks/ Python.framework/Versions/2.3/lib/python2.3/site-packages/f2py2e/src/ fortranobject.o build/temp.darwin-7.0.0-Power_Macintosh-2.3/build/temp.darwin-7.0.0- Power_Macintosh-2.3/fblasmodule.o -L/usr/local/lib -Lbuild/temp.darwin-7.0.0-Power_Macintosh-2.3 -Lbuild/temp.darwin-7.0.0-Power_Macintosh-2.3 -L/usr/local/lib/gcc/powerpc-apple-darwin6.6/3.4 -L/usr/local/lib -lfblas_f2py -llapack -lf77blas -lcblas -latlas -lc_misc -lcephes -lrootfind -lg2c -o build/lib.darwin-7.0.0-Power_Macintosh-2.3/scipy/linalg/fblas.so ld: Undefined symbols: restFP saveFP restVEC_vr10 saveVEC_vr11 error: command 'gcc' failed with exit status 1 notice I have the -faltivec flag included, but still the undefined symbols. Any ideas? Thanks (again) in advance, 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 bob at redivi.com Fri Nov 14 13:03:18 2003 From: bob at redivi.com (Bob Ippolito) Date: Fri, 14 Nov 2003 13:03:18 -0500 Subject: [SciPy-user] compile failure on OSX In-Reply-To: References: Message-ID: On Nov 11, 2003, at 8:06 AM, Christopher Fonnesbeck wrote: > I now have wxPython working on OSX 10.3, but the scipy build still > fails. Here is the error: > > gcc -Wl,-F. -Wl,-F. -bundle -framework Python -faltivec > build/temp.darwin-7.0.0-Power_Macintosh-2.3/System/Library/Frameworks/ > Python.framework/Versions/2.3/lib/python2.3/site-packages/f2py2e/src/ > fortranobject.o > build/temp.darwin-7.0.0-Power_Macintosh-2.3/build/temp.darwin-7.0.0- > Power_Macintosh-2.3/fblasmodule.o -L/usr/local/lib > -Lbuild/temp.darwin-7.0.0-Power_Macintosh-2.3 > -Lbuild/temp.darwin-7.0.0-Power_Macintosh-2.3 > -L/usr/local/lib/gcc/powerpc-apple-darwin6.6/3.4 -L/usr/local/lib > -lfblas_f2py -llapack -lf77blas -lcblas -latlas -lc_misc -lcephes > -lrootfind -lg2c -o > build/lib.darwin-7.0.0-Power_Macintosh-2.3/scipy/linalg/fblas.so > ld: Undefined symbols: > restFP > saveFP > restVEC_vr10 > saveVEC_vr11 > error: command 'gcc' failed with exit status 1 > > notice I have the -faltivec flag included, but still the undefined > symbols. Any ideas? You seem to be using gcc 3.4, not Apple's GCC 3.3 .. I have no idea how that acts, but that's definitely at least part of your problem. I'm relatively sure restFP/etc are part of libc in Apple's distribution of GCC 3.3. -bob -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 2357 bytes Desc: not available URL: From chris at fonnesbeck.org Fri Nov 14 16:14:00 2003 From: chris at fonnesbeck.org (Christopher Fonnesbeck) Date: Fri, 14 Nov 2003 16:14:00 -0500 Subject: [SciPy-user] compile failure on OSX In-Reply-To: References: Message-ID: <77681D10-16E7-11D8-876B-000A956FDAC0@fonnesbeck.org> Not so ... I am using gcc that shipped with Panther, which is 3.3 it seems: Goldeneye:~ chris$ gcc -v Reading specs from /usr/libexec/gcc/darwin/ppc/3.3/specs Thread model: posix gcc version 3.3 20030304 (Apple Computer, Inc. build 1495) On Nov 14, 2003, at 1:03 PM, Bob Ippolito wrote: > > On Nov 11, 2003, at 8:06 AM, Christopher Fonnesbeck wrote: > >> I now have wxPython working on OSX 10.3, but the scipy build still >> fails. Here is the error: >> >> gcc -Wl,-F. -Wl,-F. -bundle -framework Python -faltivec >> build/temp.darwin-7.0.0-Power_Macintosh-2.3/System/Library/ >> Frameworks/Python.framework/Versions/2.3/lib/python2.3/site-packages/ >> f2py2e/src/fortranobject.o >> build/temp.darwin-7.0.0-Power_Macintosh-2.3/build/temp.darwin-7.0.0- >> Power_Macintosh-2.3/fblasmodule.o -L/usr/local/lib >> -Lbuild/temp.darwin-7.0.0-Power_Macintosh-2.3 >> -Lbuild/temp.darwin-7.0.0-Power_Macintosh-2.3 >> -L/usr/local/lib/gcc/powerpc-apple-darwin6.6/3.4 -L/usr/local/lib >> -lfblas_f2py -llapack -lf77blas -lcblas -latlas -lc_misc -lcephes >> -lrootfind -lg2c -o >> build/lib.darwin-7.0.0-Power_Macintosh-2.3/scipy/linalg/fblas.so >> ld: Undefined symbols: >> restFP >> saveFP >> restVEC_vr10 >> saveVEC_vr11 >> error: command 'gcc' failed with exit status 1 >> >> notice I have the -faltivec flag included, but still the undefined >> symbols. Any ideas? > > You seem to be using gcc 3.4, not Apple's GCC 3.3 .. I have no idea > how that acts, but that's definitely at least part of your problem. > I'm relatively sure restFP/etc are part of libc in Apple's > distribution of GCC 3.3. > > -bob_______________________________________________ > SciPy-user mailing list > SciPy-user at scipy.net > http://www.scipy.net/mailman/listinfo/scipy-user > -- 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 bob at redivi.com Fri Nov 14 16:28:00 2003 From: bob at redivi.com (Bob Ippolito) Date: Fri, 14 Nov 2003 16:28:00 -0500 Subject: [SciPy-user] compile failure on OSX In-Reply-To: <77681D10-16E7-11D8-876B-000A956FDAC0@fonnesbeck.org> References: <77681D10-16E7-11D8-876B-000A956FDAC0@fonnesbeck.org> Message-ID: <6C8F030D-16E9-11D8-8950-000A95686CD8@redivi.com> Then your computer is very confused: >> -L/usr/local/lib/gcc/powerpc-apple-darwin6.6/3.4 ^^^ this is some 3.4 version of a GCC product compiled for 10.2.6 (I think, 6.8 is the latest jag-darwin). This might be one of those ugly fink-isms(?). Since it's -L, it comes on the search path before even /usr/lib, so you might be linking to the libSystem from some whack non-Apple version of GCC that is missing the FPU/Altivec helper functions that your gcc 3.3 compiler expects to see. -bob On Nov 14, 2003, at 4:14 PM, Christopher Fonnesbeck wrote: > Not so ... I am using gcc that shipped with Panther, which is 3.3 it > seems: > > Goldeneye:~ chris$ gcc -v > Reading specs from /usr/libexec/gcc/darwin/ppc/3.3/specs > Thread model: posix > gcc version 3.3 20030304 (Apple Computer, Inc. build 1495) > > On Nov 14, 2003, at 1:03 PM, Bob Ippolito wrote: > >> >> On Nov 11, 2003, at 8:06 AM, Christopher Fonnesbeck wrote: >> >>> I now have wxPython working on OSX 10.3, but the scipy build still >>> fails. Here is the error: >>> >>> gcc -Wl,-F. -Wl,-F. -bundle -framework Python -faltivec >>> build/temp.darwin-7.0.0-Power_Macintosh-2.3/System/Library/ >>> Frameworks/Python.framework/Versions/2.3/lib/python2.3/site- >>> packages/f2py2e/src/fortranobject.o >>> build/temp.darwin-7.0.0-Power_Macintosh-2.3/build/temp.darwin-7.0.0- >>> Power_Macintosh-2.3/fblasmodule.o -L/usr/local/lib >>> -Lbuild/temp.darwin-7.0.0-Power_Macintosh-2.3 >>> -Lbuild/temp.darwin-7.0.0-Power_Macintosh-2.3 >>> -L/usr/local/lib/gcc/powerpc-apple-darwin6.6/3.4 -L/usr/local/lib >>> -lfblas_f2py -llapack -lf77blas -lcblas -latlas -lc_misc -lcephes >>> -lrootfind -lg2c -o >>> build/lib.darwin-7.0.0-Power_Macintosh-2.3/scipy/linalg/fblas.so >>> ld: Undefined symbols: >>> restFP >>> saveFP >>> restVEC_vr10 >>> saveVEC_vr11 >>> error: command 'gcc' failed with exit status 1 >>> >>> notice I have the -faltivec flag included, but still the undefined >>> symbols. Any ideas? >> >> You seem to be using gcc 3.4, not Apple's GCC 3.3 .. I have no idea >> how that acts, but that's definitely at least part of your problem. >> I'm relatively sure restFP/etc are part of libc in Apple's >> distribution of GCC 3.3. >> >> -bob_______________________________________________ >> SciPy-user mailing list >> SciPy-user at scipy.net >> http://www.scipy.net/mailman/listinfo/scipy-user >> > -- > 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 -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 2357 bytes Desc: not available URL: From byronea at u.washington.edu Sat Nov 15 12:44:51 2003 From: byronea at u.washington.edu (Byron Amerson) Date: Sat, 15 Nov 2003 09:44:51 -0800 (PST) Subject: [SciPy-user] compil eissues on OSX 10.3.1 Message-ID: Hello I am new to the list. I am excited about using SciPy, but have run into a speedbump in building it on my mac. In this message I have include ouput from python setup.py install, python -c 'from f2py2e.diagnose import run;run()', and g77 --version. I have setenv ATALS /sw/lib(as an aside, is it possible to build against mac's frameworks of lapack/blas??), I have wxPython for mac installed. THANKS IN ADVANCE... thus: . . . package init file '/usr/local/src/scipy/SciPy-0.2.0_alpha_200.4161/Lib/stats/tests/__init__.py' not found (or not a regular file) package init file '/usr/local/src/scipy/SciPy-0.2.0_alpha_200.4161/Lib/optimize/tests/__init__.py' not found (or not a regular file) package init file '/usr/local/src/scipy/SciPy-0.2.0_alpha_200.4161/Lib/fftpack/tests/__init__.py' not found (or not a regular file) package init file '/usr/local/src/scipy/SciPy-0.2.0_alpha_200.4161/Lib/fftpack/tests/__init__.py' not found (or not a regular file) running build_clib running run_f2py running build_flib f771: error: unrecognized command line option "-arch" f771: fatal error: can't open ppc: No such file or directory compilation terminated. error: failure during compile (exit status = 256) HERE IS THE OUTPUT OF python -c 'from f2py2e.diagnose import run;run()' : ------ os.name='posix' ------ sys.platform='darwin' ------ sys.version: 2.3 (#1, Sep 13 2003, 00:49:11) [GCC 3.3 20030304 (Apple Computer, Inc. build 1495)] ------ sys.prefix: /System/Library/Frameworks/Python.framework/Versions/2.3 ------ sys.path='/Users/tectonic:/System/Library/Frameworks/Python.framework/Versions/2.3/ lib/python23.zip:/System/Library/Frameworks/Python.framework/Versions/2.3/lib/ python2.3:/System/Library/Frameworks/Python.framework/Versions/2.3/lib/python2.3/ plat-darwin:/System/Library/Frameworks/Python.framework/Versions/2.3/lib/python2.3/ plat-mac:/System/Library/Frameworks/Python.framework/Versions/2.3/lib/python2.3/ plat-mac/lib-scriptpackages:/System/Library/Frameworks/Python.framework/Versions/ 2.3/lib/python2.3/lib-tk:/System/Library/Frameworks/Python.framework/Versions/2.3/ lib/python2.3/lib-dynload:/System/Library/ Frameworks/Python.framework/Versions/2.3/lib/ python2.3/site-packages: /System/Library/Frameworks/Python.framework/Versions/2.3/lib/ python2.3/site-packages/Numeric' ------ Found Numeric version '23.1' in /System/Library/Frameworks/Python.framework/Versions/2.3/lib/python2.3/site-packages /Numeric/Numeric.pyc ------ Found f2py2e version '2.37.233-1545' in /System/Library/Frameworks/Python.framework/Versions/2.3/lib/python2.3/site-packages /f2py2e/f2py2e.pyc ------ Found scipy_distutils version '0.2.0_alpha_3.294' in '/System/Library/Frameworks/Python.framework/Versions/2.3/lib/python2.3/site-packages /scipy_distutils/__init__.pyc' Importing scipy_distutils.command.build_flib ... ok ------ Checking availability of supported Fortran compilers: ------ Importing scipy_distutils.command.cpuinfo ... ok CPU information: getNCPUs is_ppc is_ppc7450 ------ AND Tectonic:~] tectonic% g77 --version GNU Fortran (GCC) 3.4 20031015 (experimental) Copyright (C) 2002 Free Software Foundation, Inc. GNU Fortran comes with NO WARRANTY, to the extent permitted by law. You may redistribute copies of GNU Fortran under the terms of the GNU General Public License. For more information about these matters, see the file named COPYING or type the command `info -f g77 Copying'. the above was installed via fink - I ma not sure what/if mac comes installed with a fortran compiler - I looked, but not too hard... Byron Amerson MS student University of Washington Department of Earth and Space Sciences From pearu at scipy.org Sat Nov 15 16:38:37 2003 From: pearu at scipy.org (Pearu Peterson) Date: Sat, 15 Nov 2003 15:38:37 -0600 (CST) Subject: [SciPy-user] compil eissues on OSX 10.3.1 In-Reply-To: Message-ID: On Sat, 15 Nov 2003, Byron Amerson wrote: > Hello I am new to the list. > I am excited about using SciPy, but have run into a speedbump in building > it on my mac. In this message I have include ouput from python setup.py > install, python -c 'from f2py2e.diagnose import run;run()', and g77 > --version. I have setenv ATALS /sw/lib(as an aside, is it possible to > build against mac's frameworks of lapack/blas?? Yes, then you have to specify BLAS and LAPACK environment variables that should contain the directories of blas and lapack libraries, respectively. See scipy_distutils/system_info.py for more information. >), I have wxPython for mac > installed. THANKS IN ADVANCE... > running build_flib > f771: error: unrecognized command line option "-arch" > f771: fatal error: can't open ppc: No such file or directory > compilation terminated. > error: failure during compile (exit status = 256) This is fixed in recent scipy_distutils CVS. Pearu From bob at redivi.com Sat Nov 15 16:53:08 2003 From: bob at redivi.com (Bob Ippolito) Date: Sat, 15 Nov 2003 16:53:08 -0500 Subject: [SciPy-user] compil eissues on OSX 10.3.1 In-Reply-To: References: Message-ID: <19731818-17B6-11D8-82BC-000A95686CD8@redivi.com> On Nov 15, 2003, at 4:38 PM, Pearu Peterson wrote: > > > On Sat, 15 Nov 2003, Byron Amerson wrote: > >> Hello I am new to the list. >> I am excited about using SciPy, but have run into a speedbump in >> building >> it on my mac. In this message I have include ouput from python >> setup.py >> install, python -c 'from f2py2e.diagnose import run;run()', and g77 >> --version. I have setenv ATALS /sw/lib(as an aside, is it possible to >> build against mac's frameworks of lapack/blas?? > > Yes, then you have to specify BLAS and LAPACK environment variables > that should contain the directories of blas and lapack libraries, > respectively. See scipy_distutils/system_info.py for more information. I'm not sure if the LAPACK/BLAS in vecLib is enough for SciPy. From what I remember, the BLAS symbols are private to the framework, and the LAPACK in there may only be for C. This was with 10.2 though, I will try again with 10.3 sometime soon. -bob -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 2357 bytes Desc: not available URL: From byronea at u.washington.edu Tue Nov 18 00:14:32 2003 From: byronea at u.washington.edu (Byron Amerson) Date: Mon, 17 Nov 2003 21:14:32 -0800 (PST) Subject: [SciPy-user] Re: compile issues on OSX 10.3.1 In-Reply-To: References: Message-ID: Thanks for your response to my last post Pearu. First of all I'd like to offer my appologies for all of the typos in my last post! OK, I tried the a cvs from Saturday afternoon, it did not work. I'll post the output below, but some questions first. A few things setup.py is looking for, fftw, dfftw, and djbfft, I have installed. I set the environment variables like so : setenv FFTW /sw/lib *this is version2.1.3* setenv DJBFFT /usr/local/djbfft Why does setup.py not see these? See output below... Reading through the INSTALL.txt file I run across the bit: 9) chaco.tkplt requires Debian packages: python-opengl, python-pmw So I must use dselect or aptget to aquire and install these packages? Or is there another way? I have dselect and aptget as installed by/for fink, can I use them to install Debian packages?? Thanks in advance. HERE'S THE OUTPUT(I've cut out the middle portion): [Tectonic:local/src/scipy] root# python setup.py install fftw_info: NOT AVAILABLE dfftw_info: NOT AVAILABLE FFTW (http://www.fftw.org/) libraries not found. Directories to search for the libraries can be specified in the scipy_distutils/site.cfg file (section [fftw]) or by setting the FFTW environment variable. djbfft_info: NOT AVAILABLE DJBFFT (http://cr.yp.to/djbfft.html) libraries not found. Directories to search for the libraries can be specified in the scipy_distutils/site.cfg file (section [djbfft]) or by setting the DJBFFT environment variable. atlas_info: FOUND: libraries = ['lapack', 'f77blas', 'cblas', 'atlas'] library_dirs = ['/sw/lib'] creating build/temp.darwin-7.0.0-Power_Macintosh-2.3 creating build/temp.darwin-7.0.0-Power_Macintosh-2.3/linalg creating build/temp.darwin-7.0.0-Power_Macintosh-2.3/linalg/atlas_version /usr/bin/python /usr/local/src/scipy/Lib/linalg/setup_atlas_version.py build_ext --inplace --force /usr/bin/python -c "import atlas_version" ATLAS version 3.4.1 creating build/temp.darwin-7.0.0-Power_Macintosh-2.3/config_pygist ============= begin top level configuration ============= ***LOTS OF OUTPUT HERE I"LL SKIP*** running install running build running build_py package init file 'Lib/tests/__init__.py' not found (or not a regular file) package init file '/usr/local/src/scipy/scipy_core/scipy_base/tests/__init__.py' not found (or not a regular file) package init file '/usr/local/src/scipy/Lib/cluster/tests/__init__.py' not found (or not a regular file) package init file '/usr/local/src/scipy/Lib/cow/tests/__init__.py' not found (or not a regular file) package init file '/usr/local/src/scipy/Lib/fftpack/tests/__init__.py' not found (or not a regular file) package init file '/usr/local/src/scipy/Lib/fftpack/tests/__init__.py' not found (or not a regular file) package init file '/usr/local/src/scipy/Lib/gui_thread/tests/__init__.py' not found (or not a regular file) package init file '/usr/local/src/scipy/Lib/interpolate/tests/__init__.py' not found (or not a regular file) package init file '/usr/local/src/scipy/Lib/io/tests/__init__.py' not found (or not a regular file) package init file '/usr/local/src/scipy/Lib/linalg/tests/__init__.py' not found (or not a regular file) package init file '/usr/local/src/scipy/Lib/optimize/tests/__init__.py' not found (or not a regular file) package init file '/usr/local/src/scipy/Lib/signal/tests/__init__.py' not found (or not a regular file) package init file '/usr/local/src/scipy/Lib/special/tests/__init__.py' not found (or not a regular file) package init file '/usr/local/src/scipy/Lib/stats/tests/__init__.py' not found (or not a regular file) package init file 'Lib/weave/tests/__init__.py' not found (or not a regular file) package init file 'Lib/weave/tests/__init__.py' not found (or not a regular file) package init file '/usr/local/src/scipy/Lib/xxx/tests/__init__.py' not found (or not a regular file) package init file '/usr/local/src/scipy/Lib/xxx/yyy/tests/__init__.py' not found (or not a regular file) package init file '/usr/local/src/scipy/Lib_chaco/chaco/tests/__init__.py' not found (or not a regular file) package init file '/usr/local/src/scipy/Lib_chaco/kiva/tests/__init__.py' not found (or not a regular file) package init file '/usr/local/src/scipy/Lib_chaco/kiva/tests/__init__.py' not found (or not a regular file) package init file '/usr/local/src/scipy/Lib_chaco/kiva/agg/swig_src/tests/__init__.py' not found (or not a regular file) package init file '/usr/local/src/scipy/Lib_chaco/traits/tests/__init__.py' not found (or not a regular file) package init file '/usr/local/src/scipy/Lib_chaco/freetype/tests/__init__.py' not found (or not a regular file) package init file 'Lib/tests/__init__.py' not found (or not a regular file) package init file '/usr/local/src/scipy/scipy_core/scipy_base/tests/__init__.py' not found (or not a regular file) package init file '/usr/local/src/scipy/Lib/cluster/tests/__init__.py' not found (or not a regular file) package init file '/usr/local/src/scipy/Lib/cow/tests/__init__.py' not found (or not a regular file) package init file '/usr/local/src/scipy/Lib/fftpack/tests/__init__.py' not found (or not a regular file) package init file '/usr/local/src/scipy/Lib/fftpack/tests/__init__.py' not found (or not a regular file) package init file '/usr/local/src/scipy/Lib/gui_thread/tests/__init__.py' not found (or not a regular file) package init file '/usr/local/src/scipy/Lib/interpolate/tests/__init__.py' not found (or not a regular file) package init file '/usr/local/src/scipy/Lib/io/tests/__init__.py' not found (or not a regular file) package init file '/usr/local/src/scipy/Lib/linalg/tests/__init__.py' not found (or not a regular file) package init file '/usr/local/src/scipy/Lib/optimize/tests/__init__.py' not found (or not a regular file) package init file '/usr/local/src/scipy/Lib/signal/tests/__init__.py' not found (or not a regular file) package init file '/usr/local/src/scipy/Lib/special/tests/__init__.py' not found (or not a regular file) package init file '/usr/local/src/scipy/Lib/stats/tests/__init__.py' not found (or not a regular file) package init file 'Lib/weave/tests/__init__.py' not found (or not a regular file) package init file 'Lib/weave/tests/__init__.py' not found (or not a regular file) package init file '/usr/local/src/scipy/Lib/xxx/tests/__init__.py' not found (or not a regular file) package init file '/usr/local/src/scipy/Lib/xxx/yyy/tests/__init__.py' not found (or not a regular file) package init file '/usr/local/src/scipy/Lib_chaco/chaco/tests/__init__.py' not found (or not a regular file) package init file '/usr/local/src/scipy/Lib_chaco/kiva/tests/__init__.py' not found (or not a regular file) package init file '/usr/local/src/scipy/Lib_chaco/kiva/tests/__init__.py' not found (or not a regular file) package init file '/usr/local/src/scipy/Lib_chaco/kiva/agg/swig_src/tests/__init__.py' not found (or not a regular file) package init file '/usr/local/src/scipy/Lib_chaco/traits/tests/__init__.py' not found (or not a regular file) package init file '/usr/local/src/scipy/Lib_chaco/freetype/tests/__init__.py' not found (or not a regular file) running build_clib building 'rootfind' library creating build/temp.darwin-7.0.0-Power_Macintosh-2.3/Lib creating build/temp.darwin-7.0.0-Power_Macintosh-2.3/Lib/optimize creating build/temp.darwin-7.0.0-Power_Macintosh-2.3/Lib/optimize/Zeros gcc -fno-strict-aliasing -Wno-long-double -no-cpp-precomp -mno-fused-madd -fno-common -dynamic -DNDEBUG -g -O3 -Wstrict-prototypes -c Lib/optimize/Zeros/brentq.c -o build/temp.darwin-7.0.0-Power_Macintosh-2.3/Lib/optimize/Zeros/brentq.o gcc -fno-strict-aliasing -Wno-long-double -no-cpp-precomp -mno-fused-madd -fno-common -dynamic -DNDEBUG -g -O3 -Wstrict-prototypes -c Lib/optimize/Zeros/ridder.c -o build/temp.darwin-7.0.0-Power_Macintosh-2.3/Lib/optimize/Zeros/ridder.o gcc -fno-strict-aliasing -Wno-long-double -no-cpp-precomp -mno-fused-madd -fno-common -dynamic -DNDEBUG -g -O3 -Wstrict-prototypes -c Lib/optimize/Zeros/bisect.c -o build/temp.darwin-7.0.0-Power_Macintosh-2.3/Lib/optimize/Zeros/bisect.o gcc -fno-strict-aliasing -Wno-long-double -no-cpp-precomp -mno-fused-madd -fno-common -dynamic -DNDEBUG -g -O3 -Wstrict-prototypes -c Lib/optimize/Zeros/brenth.c -o build/temp.darwin-7.0.0-Power_Macintosh-2.3/Lib/optimize/Zeros/brenth.o ar -cr build/temp.darwin-7.0.0-Power_Macintosh-2.3/librootfind.a build/temp.darwin-7.0.0-Power_Macintosh-2.3/Lib/optimize/Zeros/bisect.o build/temp.darwin-7.0.0-Power_Macintosh-2.3/Lib/optimize/Zeros/brenth.o build/temp.darwin-7.0.0-Power_Macintosh-2.3/Lib/optimize/Zeros/brentq.o build/temp.darwin-7.0.0-Power_Macintosh-2.3/Lib/optimize/Zeros/ridder.o ranlib build/temp.darwin-7.0.0-Power_Macintosh-2.3/librootfind.a building 'superlu' library creating build/temp.darwin-7.0.0-Power_Macintosh-2.3/Lib/sparse creating build/temp.darwin-7.0.0-Power_Macintosh-2.3/Lib/sparse/SuperLU2.0 creating build/temp.darwin-7.0.0-Power_Macintosh-2.3/Lib/sparse/SuperLU2.0/SRC gcc -fno-strict-aliasing -Wno-long-double -no-cpp-precomp -mno-fused-madd -fno-common -dynamic -DNDEBUG -g -O3 -Wstrict-prototypes -I/System/Library/Frameworks/Python.framework/Versions/2.3/include/python2.3 -c Lib/sparse/SuperLU2.0/SRC/zutil.c -o build/temp.darwin-7.0.0-Power_Macintosh-2.3/Lib/sparse/SuperLU2.0/SRC/zutil.o In file included from Lib/sparse/SuperLU2.0/SRC/zutil.c:25: Lib/sparse/SuperLU2.0/SRC/util.h:7:20: malloc.h: No such file or directory In file included from Lib/sparse/SuperLU2.0/SRC/zutil.c:25: Lib/sparse/SuperLU2.0/SRC/util.h:95: warning: function declaration isn't a prototype Lib/sparse/SuperLU2.0/SRC/util.h:101: warning: function declaration isn't a prototype error: command 'gcc' failed with exit status 1 Byron Amerson MS student University of Washington Department of Earth and Space Sciences From pearu at scipy.org Tue Nov 18 09:19:24 2003 From: pearu at scipy.org (Pearu Peterson) Date: Tue, 18 Nov 2003 08:19:24 -0600 (CST) Subject: [SciPy-user] Re: compile issues on OSX 10.3.1 In-Reply-To: Message-ID: On Mon, 17 Nov 2003, Byron Amerson wrote: > OK, I tried the a cvs from Saturday afternoon, it did not work. I'll > post the output below, but some questions first. > > A few things setup.py is looking for, fftw, dfftw, and djbfft, I have > installed. I set the environment variables like so : > > setenv FFTW /sw/lib *this is version2.1.3* What is the fftw related contents of /sw/lib directory? Where are fftw header files installed? > setenv DJBFFT /usr/local/djbfft Try also setenv DJBFFT /usr/local > Why does setup.py not see these? See output below... Run python scipy_core/scipy_distutils/system_info.py to see more verbose output (showing what files/directories are tested for finding software) > Reading through the INSTALL.txt file I run across the bit: > > 9) chaco.tkplt requires > Debian packages: python-opengl, python-pmw > > So I must use dselect or aptget to aquire and install these packages? Or > is there another way? I have dselect and aptget as installed by/for fink, > can I use them to install Debian packages?? You can try, I have no idea if it works. On the other hand, chaco is broken for all platforms except Windows. So, at the moment you can (should) ignore chaco. > package init file 'Lib/tests/__init__.py' not found (or not a regular > file) These messages about not finding tests/__init__.py files are normal. > build/temp.darwin-7.0.0-Power_Macintosh-2.3/Lib/sparse/SuperLU2.0/SRC > gcc -fno-strict-aliasing -Wno-long-double -no-cpp-precomp -mno-fused-madd > -fno-common -dynamic -DNDEBUG -g -O3 -Wstrict-prototypes > -I/System/Library/Frameworks/Python.framework/Versions/2.3/include/python2.3 > -c Lib/sparse/SuperLU2.0/SRC/zutil.c -o > build/temp.darwin-7.0.0-Power_Macintosh-2.3/Lib/sparse/SuperLU2.0/SRC/zutil.o > In file included from Lib/sparse/SuperLU2.0/SRC/zutil.c:25: > Lib/sparse/SuperLU2.0/SRC/util.h:7:20: malloc.h: No such file or directory Hmm, could you find out why malloc.h is not found? Is it in your system and where? Pearu From prabhu at aero.iitm.ernet.in Tue Nov 18 14:01:53 2003 From: prabhu at aero.iitm.ernet.in (Prabhu Ramachandran) Date: Wed, 19 Nov 2003 00:31:53 +0530 Subject: [SciPy-user] ANN: MayaVi-1.3 released Message-ID: <16314.27809.836219.483395@monster.linux.in> Hi, This is to announce the availability of the MayaVi Data Visualizer version 1.3. MayaVi is a free, easy to use, scientific data visualizer. It is written in Python and uses the Visualization Toolkit (VTK) for the graphics. It provides a GUI written using Tkinter. MayaVi is free and distributed under a BSD license. It is also cross platform and should run on any platform where both Python and VTK are available. For more information, sources, binaries, screenshots, installation instructions, documentation etc. visit the MayaVi home page at: http://mayavi.sourceforge.net Also bundled with MayaVi is a VTK pipeline browser written in Python and a utility module that makes using VTK easier from the Python interpreter. New in this release: * New Modules: - Volume: Volume visualization using either texture or ray cast mappers. A useful GUI color transfer function editor is also built-in. - Glyph: Generic glyphs for scalars and vectors. - BandedSurfaceMap: Display banded contours of input data. - Labels: Automatic labeling of data used in other modules. - Text: Display 2D text strings. * New Filters: - ExtractGrid: Select part of a grid or subsample it. - StructuredPointsProbe: Probes any input data onto a structured points grid so the data can be used for Volume visualization. - Delaunay3D/Delaunay2D: Triangulate input point data. - UserDefined: Manage any user specified VTK filter. - CutPlane: Cut through any input dataset with a plane. * Support for VTK's new XML based files and for EnSight case files (EnSight6 and EnSightGold formats are supported). * Improved VectorCutPlane and VelocityVector so you can use a 3D arrow and any of the standard 2D glyphs. * The 'imv' module is now part of the mayavi.tools package. imv is a simple MayaVi based module that provides Matlab like one liners to view 2D surfaces and arrays from the Python interpreter. * Support to save rendered scene to vector EPS/PS/PDF or save the text to TeX output using GL2PS. Requires VTK to be built with GL2PS support (the MayaVi binaries support this). Also added support for exporting to a Wavefront OBJ file. * Support to save the stereo rendering options, the lookup table settings and the lighting as defaults. * Default lookup tables have better contrast. 28 user contributed lookup tables are also included in the sources. Any lookup table can be reversed. * Improved vtkPipeline modules: - ConfigVtkObj: Separated functionality of the ConfigVtkObj class into a separate ConfigVtkObjFrame and a driver class. More options to generate GUI's automatically. - VtkPickler: Can save/load state from a dictionary. * Fixes for Python-2.3 and Tk-8.4 related changes. * Fixes for the Mac OS X. * Many other fixes and improvements. Acknowledgements: Many thanks to the following folks. - Gerard Gorman: initial volume module with GUI CTF editor. - Axel Steuwer: initial Delaunay3D filter. - Richard Boardman: patch for OBJ export. - Fernando Perez and Arnd Baecker: user contributed lookup tables. - P. N. Badri Narayanan: Win32 VTK DLL's. - Thanks also to others who reported bugs and sent in bug fixes. Have fun! prabhu From sluis at eden.rutgers.edu Fri Nov 21 12:07:31 2003 From: sluis at eden.rutgers.edu (Arend SLUIS) Date: Fri, 21 Nov 2003 12:07:31 -0500 (EST) Subject: [SciPy-user] cephes -> scipy Message-ID: Hi, I am using some python software that relies on the cephes special function package. Since I am unable to compile cephes on my machine, I have been thinking of hacking the code and switching to scipy's suite of special functions. Does anyone have any experience with both cephes and scipy? Any comments and suggestions are appreciated. Arend. *Newbie in all things python* From oliphant at ee.byu.edu Fri Nov 21 15:35:15 2003 From: oliphant at ee.byu.edu (Travis E. Oliphant) Date: Fri, 21 Nov 2003 13:35:15 -0700 Subject: [SciPy-user] cephes -> scipy In-Reply-To: References: Message-ID: <3FBE7703.9060605@ee.byu.edu> Arend SLUIS wrote: > Hi, > > I am using some python software that relies on the cephes special > function package. Since I am unable to compile cephes on my machine, I > have been thinking of hacking the code and switching to scipy's suite of > special functions. Does anyone have any experience with both cephes and > scipy? Any comments and suggestions are appreciated. All of the cephes functions are in scipy.special. Much more is there as well. You should be able to do a search replace from cephes -> scipy.special to port your code. There may be a couple of name changes but I don't think so. If you already have ATLAS or *complete* LAPACK/BLAS then installing SciPy is not difficult. Check the latest CVS out and run python setup.py install. If you are on Windows, then download Enthought's edition to avoid compilation of the package. -Travis O. From travis at enthought.com Fri Nov 21 15:52:25 2003 From: travis at enthought.com (Travis N. Vaught) Date: Fri, 21 Nov 2003 14:52:25 -0600 Subject: [SciPy-user] cephes -> scipy In-Reply-To: <3FBE7703.9060605@ee.byu.edu> Message-ID: <002901c3b071$63425370$0200a8c0@tvlaptop> > -----Original Message----- > From: scipy-user-bounces at scipy.net [mailto:scipy-user-bounces at scipy.net] > On Behalf Of Travis E. Oliphant > Sent: Friday, November 21, 2003 2:35 PM > To: SciPy Users List > Subject: Re: [SciPy-user] cephes -> scipy > > If you already have ATLAS or *complete* LAPACK/BLAS then installing > SciPy is not difficult. > > Check the latest CVS out and run python setup.py install. > > If you are on Windows, then download Enthought's edition to avoid > compilation of the package. > > -Travis O. > Python 2.3.2 for Windows - Enthought Edition may be found here: http://www.enthought.com/python From cjw at sympatico.ca Mon Nov 24 08:17:04 2003 From: cjw at sympatico.ca (Colin J. Williams) Date: Mon, 24 Nov 2003 08:17:04 -0500 Subject: [SciPy-user] PyMatrix: Announcement Message-ID: <3FC204D0.2010503@sympatico.ca> PyMatrix is available for test and review. http://www3.sympatico.ca/cjw PyMatrix provides access to basic matrix arithmetic, using Python and numarray. Examples: A * B => the product of matrices A and B A.I => the inverse of matrix A A.EVectors => the eigenvectors of A A.var(0) => the variances of the columns of A (a.T*a).I * a.T*b => the solution (x) for a * x = b, where a is a matrix and b a column vector This package was developed on a Windows XP. I would appreciate comments, particularly with respect to usage on other systems. Colin W. From chris at fonnesbeck.org Mon Nov 24 09:10:22 2003 From: chris at fonnesbeck.org (Christopher Fonnesbeck) Date: Mon, 24 Nov 2003 09:10:22 -0500 Subject: [SciPy-user] PyMatrix: Announcement In-Reply-To: <3FC204D0.2010503@sympatico.ca> References: <3FC204D0.2010503@sympatico.ca> Message-ID: I only see the windows installer; is there a tar or zip archive of the modules? C. On Nov 24, 2003, at 8:17 AM, Colin J. Williams wrote: > PyMatrix is available for test and review. > http://www3.sympatico.ca/cjw > > PyMatrix provides access to basic matrix arithmetic, using Python and > numarray. > > Examples: > A * B => the product of > matrices A and B > A.I => the inverse of > matrix A > A.EVectors => the eigenvectors of A > A.var(0) => the variances of the > columns of A > (a.T*a).I * a.T*b => the solution (x) for a * > x = b, > where a is a > matrix and b a column vector > > This package was developed on a Windows XP. I would appreciate > comments, particularly with respect to usage on other systems. > > Colin W. > > > > _______________________________________________ > SciPy-user mailing list > SciPy-user at scipy.net > http://www.scipy.net/mailman/listinfo/scipy-user > > -- 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 nadavh at VisionSense.com Mon Nov 24 09:28:22 2003 From: nadavh at VisionSense.com (Nadav Horesh) Date: 24 Nov 2003 16:28:22 +0200 Subject: [SciPy-user] Re: [Numpy-discussion] PyMatrix: Announcement In-Reply-To: <3FC204D0.2010503@sympatico.ca> References: <3FC204D0.2010503@sympatico.ca> Message-ID: <1069684102.5532.1.camel@Nadav.Envision.co.il> Don't see how can I get the sources for gnu/linux installation. Nadav On Mon, 2003-11-24 at 15:17, Colin J. Williams wrote: > PyMatrix is available for test and review. > http://www3.sympatico.ca/cjw > > PyMatrix provides access to basic matrix arithmetic, using Python and > numarray. > > Examples: > A * B => the product of > matrices A and B > A.I => the inverse of matrix A > A.EVectors => the eigenvectors of A > A.var(0) => the variances of the > columns of A > (a.T*a).I * a.T*b => the solution (x) for a * > x = b, > where a is a > matrix and b a column vector > > This package was developed on a Windows XP. I would appreciate > comments, particularly with respect to usage on other systems. > > Colin W. > > > > > > ------------------------------------------------------- > This SF.net email is sponsored by: SF.net Giveback Program. > Does SourceForge.net help you be more productive? Does it > help you create better code? SHARE THE LOVE, and help us help > YOU! Click Here: http://sourceforge.net/donate/ > _______________________________________________ > Numpy-discussion mailing list > Numpy-discussion at lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/numpy-discussion From chris at fonnesbeck.org Mon Nov 24 22:15:36 2003 From: chris at fonnesbeck.org (Christopher Fonnesbeck) Date: Mon, 24 Nov 2003 22:15:36 -0500 Subject: [SciPy-user] more OSX build problems Message-ID: I have yet to build scipy successfully on OSX, despite several months of trying, but perhaps someone is interested in the bug reports. Building the current cvs gives the following: In file included from Lib/sparse/SuperLU2.0/SRC/zutil.c:25: Lib/sparse/SuperLU2.0/SRC/util.h:7:20: malloc.h: No such file or directory In file included from Lib/sparse/SuperLU2.0/SRC/zutil.c:25: Lib/sparse/SuperLU2.0/SRC/util.h:95: warning: function declaration isn't a prototype Lib/sparse/SuperLU2.0/SRC/util.h:101: warning: function declaration isn't a prototype Lib/sparse/SuperLU2.0/SRC/zutil.c: In function `zPrint_SuperNode_Matrix': Lib/sparse/SuperLU2.0/SRC/zutil.c:245: warning: operation on `d' may be undefined Lib/sparse/SuperLU2.0/SRC/zutil.c: At top level: Lib/sparse/SuperLU2.0/SRC/zutil.c:474: warning: return type defaults to `int' error: command 'gcc' failed with exit status 1 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 oliphant at ee.byu.edu Mon Nov 24 23:00:32 2003 From: oliphant at ee.byu.edu (Travis E. Oliphant) Date: Mon, 24 Nov 2003 21:00:32 -0700 Subject: [SciPy-user] more OSX build problems In-Reply-To: References: Message-ID: <3FC2D3E0.8040101@ee.byu.edu> Christopher Fonnesbeck wrote: > I have yet to build scipy successfully on OSX, despite several months of > trying, but perhaps someone is interested in the bug reports. Building > the current cvs gives the following: > > In file included from Lib/sparse/SuperLU2.0/SRC/zutil.c:25: > Lib/sparse/SuperLU2.0/SRC/util.h:7:20: malloc.h: No such file or directory This is the problem for you. Why don't you have a malloc.h file on your system. Should this be another file for OSX? Can anybody who knows OSX help here? You can also just not compile sparse support for now. Edit the setup.py file and comment out the line which includes the sparse package. -Travis O. From bob at redivi.com Mon Nov 24 23:34:51 2003 From: bob at redivi.com (Bob Ippolito) Date: Mon, 24 Nov 2003 23:34:51 -0500 Subject: [SciPy-user] more OSX build problems In-Reply-To: <3FC2D3E0.8040101@ee.byu.edu> References: <3FC2D3E0.8040101@ee.byu.edu> Message-ID: On Nov 24, 2003, at 11:00 PM, Travis E. Oliphant wrote: > Christopher Fonnesbeck wrote: >> I have yet to build scipy successfully on OSX, despite several months >> of trying, but perhaps someone is interested in the bug reports. >> Building the current cvs gives the following: >> In file included from Lib/sparse/SuperLU2.0/SRC/zutil.c:25: >> Lib/sparse/SuperLU2.0/SRC/util.h:7:20: malloc.h: No such file or >> directory > > This is the problem for you. Why don't you have a malloc.h file on > your system. Should this be another file for OSX? > > Can anybody who knows OSX help here? > > You can also just not compile sparse support for now. Edit the > setup.py file and comment out the line which includes the sparse > package. malloc.h is not a standard header.. on OS X it's , but you shouldn't need it if you have -bob -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 2357 bytes Desc: not available URL: From pearu at scipy.org Tue Nov 25 02:58:24 2003 From: pearu at scipy.org (Pearu Peterson) Date: Tue, 25 Nov 2003 01:58:24 -0600 (CST) Subject: [SciPy-user] more OSX build problems In-Reply-To: Message-ID: On Mon, 24 Nov 2003, Bob Ippolito wrote: > > On Nov 24, 2003, at 11:00 PM, Travis E. Oliphant wrote: > > > Christopher Fonnesbeck wrote: > >> I have yet to build scipy successfully on OSX, despite several months > >> of trying, but perhaps someone is interested in the bug reports. > >> Building the current cvs gives the following: > >> In file included from Lib/sparse/SuperLU2.0/SRC/zutil.c:25: > >> Lib/sparse/SuperLU2.0/SRC/util.h:7:20: malloc.h: No such file or > >> directory > > > > This is the problem for you. Why don't you have a malloc.h file on > > your system. Should this be another file for OSX? > > > > Can anybody who knows OSX help here? > > > > You can also just not compile sparse support for now. Edit the > > setup.py file and comment out the line which includes the sparse > > package. > > malloc.h is not a standard header.. on OS X it's , but > you shouldn't need it if you have Thanks for this information. I have disabled malloc.h includes in scipy CVS. Hope it builds now under OSX. Thanks, Pearu From pearu at scipy.org Tue Nov 25 03:04:46 2003 From: pearu at scipy.org (Pearu Peterson) Date: Tue, 25 Nov 2003 02:04:46 -0600 (CST) Subject: [SciPy-user] more OSX build problems In-Reply-To: <3FC2D3E0.8040101@ee.byu.edu> Message-ID: On Mon, 24 Nov 2003, Travis E. Oliphant wrote: > You can also just not compile sparse support for now. Edit the setup.py > file and comment out the line which includes the sparse package. With the recent CVS one can disable Scipy modules by adding their names to ignore_packages list at the end of setup.py file. Or create a customized setup file similar to setup_no_chaco.py. For example, create setup_osx.py file containing: from setup import setup_package ignore_packages = ['chaco','kiva','freetype','traits', 'sparse'] if __name__ == "__main__": setup_package(ignore_packages) and then use it instead of setup.py. Pearu From chris at fonnesbeck.org Tue Nov 25 08:57:32 2003 From: chris at fonnesbeck.org (Christopher Fonnesbeck) Date: Tue, 25 Nov 2003 08:57:32 -0500 Subject: [SciPy-user] more OSX build problems In-Reply-To: References: Message-ID: <50A6CEDA-1F4F-11D8-A6C7-000A956FDAC0@fonnesbeck.org> Thanks Pearu, it seems to get beyond that now, but choking on some undefined symbols further along: gcc -Wl,-F. -Wl,-F. -bundle -framework Python -faltivec build/temp.darwin-7.0.0-Power_Macintosh-2.3/Lib/fftpack/src/zfft.o build/temp.darwin-7.0.0-Power_Macintosh-2.3/Lib/fftpack/src/drfft.o build/temp.darwin-7.0.0-Power_Macintosh-2.3/Lib/fftpack/src/zrfft.o build/temp.darwin-7.0.0-Power_Macintosh-2.3/Lib/fftpack/src/zfftnd.o build/temp.darwin-7.0.0-Power_Macintosh-2.3/System/Library/Frameworks/ Python.framework/Versions/2.3/lib/python2.3/site-packages/f2py2e/src/ fortranobject.o build/temp.darwin-7.0.0-Power_Macintosh-2.3/build/temp.darwin-7.0.0- Power_Macintosh-2.3/_fftpackmodule.o -L/usr/local/lib -Lbuild/temp.darwin-7.0.0-Power_Macintosh-2.3 -Lbuild/temp.darwin-7.0.0-Power_Macintosh-2.3 -L/usr/local/lib/gcc/powerpc-apple-darwin6.6/3.4 -L/usr/local/lib -ldfftpack -lrfftw -lfftw -lrootfind -lsuperlu -lmyblas -lc_misc -lcephes -lg2c -o build/lib.darwin-7.0.0-Power_Macintosh-2.3/scipy/fftpack/_fftpack.so ld: Undefined symbols: restFP saveFP error: command 'gcc' failed with exit status 1 I have tried using the -faltivec flag, but with the same result. On Nov 25, 2003, at 3:04 AM, Pearu Peterson wrote: > > > On Mon, 24 Nov 2003, Travis E. Oliphant wrote: > >> You can also just not compile sparse support for now. Edit the >> setup.py >> file and comment out the line which includes the sparse package. > > With the recent CVS one can disable Scipy modules by adding their > names to ignore_packages list at the end of setup.py file. > Or create a customized setup file similar to setup_no_chaco.py. > For example, create setup_osx.py file containing: > > from setup import setup_package > ignore_packages = ['chaco','kiva','freetype','traits', 'sparse'] > if __name__ == "__main__": > setup_package(ignore_packages) > > and then use it instead of setup.py. > > Pearu > > _______________________________________________ > SciPy-user mailing list > SciPy-user at scipy.net > http://www.scipy.net/mailman/listinfo/scipy-user > > -- 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 pearu at scipy.org Tue Nov 25 09:20:23 2003 From: pearu at scipy.org (Pearu Peterson) Date: Tue, 25 Nov 2003 08:20:23 -0600 (CST) Subject: [SciPy-user] more OSX build problems In-Reply-To: <50A6CEDA-1F4F-11D8-A6C7-000A956FDAC0@fonnesbeck.org> Message-ID: On Tue, 25 Nov 2003, Christopher Fonnesbeck wrote: > Thanks Pearu, it seems to get beyond that now, but choking on some > undefined symbols further along: > > gcc -Wl,-F. -Wl,-F. -bundle -framework Python -faltivec > ld: Undefined symbols: > restFP > saveFP > error: command 'gcc' failed with exit status 1 Hmm, these symbols should be OSX related. Anyone having ideas how to fix it? Try -dynamiclib option (that I pick up from google search). Pearu From bob at redivi.com Tue Nov 25 09:25:52 2003 From: bob at redivi.com (Bob Ippolito) Date: Tue, 25 Nov 2003 09:25:52 -0500 Subject: [SciPy-user] more OSX build problems In-Reply-To: <50A6CEDA-1F4F-11D8-A6C7-000A956FDAC0@fonnesbeck.org> References: <50A6CEDA-1F4F-11D8-A6C7-000A956FDAC0@fonnesbeck.org> Message-ID: <4647BCDC-1F53-11D8-8763-000A95686CD8@redivi.com> What's the output of gcc -v? I'm guessing you're not using Apple's compiler, because of the -L/usr/local/lib/gcc/powerpc-apple-darwin6.6/3.4. I'd try using Apple's gcc 3.3, the one that comes with Xcode. -bob On Nov 25, 2003, at 8:57 AM, Christopher Fonnesbeck wrote: > Thanks Pearu, it seems to get beyond that now, but choking on some > undefined symbols further along: > > gcc -Wl,-F. -Wl,-F. -bundle -framework Python -faltivec > build/temp.darwin-7.0.0-Power_Macintosh-2.3/Lib/fftpack/src/zfft.o > build/temp.darwin-7.0.0-Power_Macintosh-2.3/Lib/fftpack/src/drfft.o > build/temp.darwin-7.0.0-Power_Macintosh-2.3/Lib/fftpack/src/zrfft.o > build/temp.darwin-7.0.0-Power_Macintosh-2.3/Lib/fftpack/src/zfftnd.o > build/temp.darwin-7.0.0-Power_Macintosh-2.3/System/Library/Frameworks/ > Python.framework/Versions/2.3/lib/python2.3/site-packages/f2py2e/src/ > fortranobject.o > build/temp.darwin-7.0.0-Power_Macintosh-2.3/build/temp.darwin-7.0.0- > Power_Macintosh-2.3/_fftpackmodule.o -L/usr/local/lib > -Lbuild/temp.darwin-7.0.0-Power_Macintosh-2.3 > -Lbuild/temp.darwin-7.0.0-Power_Macintosh-2.3 > -L/usr/local/lib/gcc/powerpc-apple-darwin6.6/3.4 -L/usr/local/lib > -ldfftpack -lrfftw -lfftw -lrootfind -lsuperlu -lmyblas -lc_misc > -lcephes -lg2c -o > build/lib.darwin-7.0.0-Power_Macintosh-2.3/scipy/fftpack/_fftpack.so > ld: Undefined symbols: > restFP > saveFP > error: command 'gcc' failed with exit status 1 > > I have tried using the -faltivec flag, but with the same result. > > On Nov 25, 2003, at 3:04 AM, Pearu Peterson wrote: > >> >> >> On Mon, 24 Nov 2003, Travis E. Oliphant wrote: >> >>> You can also just not compile sparse support for now. Edit the >>> setup.py >>> file and comment out the line which includes the sparse package. >> >> With the recent CVS one can disable Scipy modules by adding their >> names to ignore_packages list at the end of setup.py file. >> Or create a customized setup file similar to setup_no_chaco.py. >> For example, create setup_osx.py file containing: >> >> from setup import setup_package >> ignore_packages = ['chaco','kiva','freetype','traits', 'sparse'] >> if __name__ == "__main__": >> setup_package(ignore_packages) >> >> and then use it instead of setup.py. >> >> Pearu >> >> _______________________________________________ >> SciPy-user mailing list >> SciPy-user at scipy.net >> http://www.scipy.net/mailman/listinfo/scipy-user >> >> > -- > 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 -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 2357 bytes Desc: not available URL: From bob at redivi.com Tue Nov 25 09:28:10 2003 From: bob at redivi.com (Bob Ippolito) Date: Tue, 25 Nov 2003 09:28:10 -0500 Subject: [SciPy-user] more OSX build problems In-Reply-To: References: Message-ID: <9872A376-1F53-11D8-8763-000A95686CD8@redivi.com> On Nov 25, 2003, at 9:20 AM, Pearu Peterson wrote: > > > On Tue, 25 Nov 2003, Christopher Fonnesbeck wrote: > >> Thanks Pearu, it seems to get beyond that now, but choking on some >> undefined symbols further along: >> >> gcc -Wl,-F. -Wl,-F. -bundle -framework Python -faltivec > > > >> ld: Undefined symbols: >> restFP >> saveFP >> error: command 'gcc' failed with exit status 1 > > Hmm, these symbols should be OSX related. Anyone having ideas how to > fix > it? Try -dynamiclib option (that I pick up from google search). -dynamiclib is not the right answer. I'm pretty sure restFP and saveFP are part of crt1.o or libSystem with Apple's compilers. I think he might be using Apple's gcc 3.3, and linking against some gcc 3.4 things. -bob -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 2357 bytes Desc: not available URL: From chris at fonnesbeck.org Tue Nov 25 09:29:29 2003 From: chris at fonnesbeck.org (Christopher Fonnesbeck) Date: Tue, 25 Nov 2003 09:29:29 -0500 Subject: [SciPy-user] more OSX build problems In-Reply-To: <4647BCDC-1F53-11D8-8763-000A95686CD8@redivi.com> References: <50A6CEDA-1F4F-11D8-A6C7-000A956FDAC0@fonnesbeck.org> <4647BCDC-1F53-11D8-8763-000A95686CD8@redivi.com> Message-ID: Thats the one I'm using; its g77 that is 3.4 (taken from the hpc site), but it was apparently compiled against gcc3.3, as it comes with the following disclaimer: This package WILL WORK with the GCC 3.3 updater as provided by Apple recently. On Nov 25, 2003, at 9:25 AM, Bob Ippolito wrote: > What's the output of gcc -v? I'm guessing you're not using Apple's > compiler, because of the > -L/usr/local/lib/gcc/powerpc-apple-darwin6.6/3.4. I'd try using > Apple's gcc 3.3, the one that comes with Xcode. > > -bob > > On Nov 25, 2003, at 8:57 AM, Christopher Fonnesbeck wrote: > >> Thanks Pearu, it seems to get beyond that now, but choking on some >> undefined symbols further along: >> >> gcc -Wl,-F. -Wl,-F. -bundle -framework Python -faltivec >> build/temp.darwin-7.0.0-Power_Macintosh-2.3/Lib/fftpack/src/zfft.o >> build/temp.darwin-7.0.0-Power_Macintosh-2.3/Lib/fftpack/src/drfft.o >> build/temp.darwin-7.0.0-Power_Macintosh-2.3/Lib/fftpack/src/zrfft.o >> build/temp.darwin-7.0.0-Power_Macintosh-2.3/Lib/fftpack/src/zfftnd.o >> build/temp.darwin-7.0.0-Power_Macintosh-2.3/System/Library/ >> Frameworks/Python.framework/Versions/2.3/lib/python2.3/site-packages/ >> f2py2e/src/fortranobject.o >> build/temp.darwin-7.0.0-Power_Macintosh-2.3/build/temp.darwin-7.0.0- >> Power_Macintosh-2.3/_fftpackmodule.o -L/usr/local/lib >> -Lbuild/temp.darwin-7.0.0-Power_Macintosh-2.3 >> -Lbuild/temp.darwin-7.0.0-Power_Macintosh-2.3 >> -L/usr/local/lib/gcc/powerpc-apple-darwin6.6/3.4 -L/usr/local/lib >> -ldfftpack -lrfftw -lfftw -lrootfind -lsuperlu -lmyblas -lc_misc >> -lcephes -lg2c -o >> build/lib.darwin-7.0.0-Power_Macintosh-2.3/scipy/fftpack/_fftpack.so >> ld: Undefined symbols: >> restFP >> saveFP >> error: command 'gcc' failed with exit status 1 >> >> I have tried using the -faltivec flag, but with the same result. >> >> On Nov 25, 2003, at 3:04 AM, Pearu Peterson wrote: >> >>> >>> >>> On Mon, 24 Nov 2003, Travis E. Oliphant wrote: >>> >>>> You can also just not compile sparse support for now. Edit the >>>> setup.py >>>> file and comment out the line which includes the sparse package. >>> >>> With the recent CVS one can disable Scipy modules by adding their >>> names to ignore_packages list at the end of setup.py file. >>> Or create a customized setup file similar to setup_no_chaco.py. >>> For example, create setup_osx.py file containing: >>> >>> from setup import setup_package >>> ignore_packages = ['chaco','kiva','freetype','traits', 'sparse'] >>> if __name__ == "__main__": >>> setup_package(ignore_packages) >>> >>> and then use it instead of setup.py. >>> >>> Pearu >>> >>> _______________________________________________ >>> SciPy-user mailing list >>> SciPy-user at scipy.net >>> http://www.scipy.net/mailman/listinfo/scipy-user >>> >>> >> -- >> 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 > _______________________________________________ > SciPy-user mailing list > SciPy-user at scipy.net > http://www.scipy.net/mailman/listinfo/scipy-user > -- 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: 3337 bytes Desc: not available URL: From bob at redivi.com Tue Nov 25 10:00:08 2003 From: bob at redivi.com (Bob Ippolito) Date: Tue, 25 Nov 2003 10:00:08 -0500 Subject: [SciPy-user] more OSX build problems In-Reply-To: References: <50A6CEDA-1F4F-11D8-A6C7-000A956FDAC0@fonnesbeck.org> <4647BCDC-1F53-11D8-8763-000A95686CD8@redivi.com> Message-ID: <0FB4F862-1F58-11D8-8763-000A95686CD8@redivi.com> Yeah, but you're using Panther.. it's questionable as to whether gcc 3.4 compiled for 10.2 w/ gcc 3.3 updater will work in 10.3 with Xcode 1.0. -bob On Nov 25, 2003, at 9:29 AM, Christopher Fonnesbeck wrote: > Thats the one I'm using; its g77 that is 3.4 (taken from the hpc > site), but it was apparently compiled against gcc3.3, as it comes with > the following disclaimer: > > This package WILL WORK with the GCC 3.3 updater as provided by Apple > recently. > > On Nov 25, 2003, at 9:25 AM, Bob Ippolito wrote: > >> What's the output of gcc -v? I'm guessing you're not using Apple's >> compiler, because of the >> -L/usr/local/lib/gcc/powerpc-apple-darwin6.6/3.4. I'd try using >> Apple's gcc 3.3, the one that comes with Xcode. >> >> -bob >> >> On Nov 25, 2003, at 8:57 AM, Christopher Fonnesbeck wrote: >> >>> Thanks Pearu, it seems to get beyond that now, but choking on some >>> undefined symbols further along: >>> >>> gcc -Wl,-F. -Wl,-F. -bundle -framework Python -faltivec >>> build/temp.darwin-7.0.0-Power_Macintosh-2.3/Lib/fftpack/src/zfft.o >>> build/temp.darwin-7.0.0-Power_Macintosh-2.3/Lib/fftpack/src/drfft.o >>> build/temp.darwin-7.0.0-Power_Macintosh-2.3/Lib/fftpack/src/zrfft.o >>> build/temp.darwin-7.0.0-Power_Macintosh-2.3/Lib/fftpack/src/zfftnd.o >>> build/temp.darwin-7.0.0-Power_Macintosh-2.3/System/Library/ >>> Frameworks/Python.framework/Versions/2.3/lib/python2.3/site- >>> packages/f2py2e/src/fortranobject.o >>> build/temp.darwin-7.0.0-Power_Macintosh-2.3/build/temp.darwin-7.0.0- >>> Power_Macintosh-2.3/_fftpackmodule.o -L/usr/local/lib >>> -Lbuild/temp.darwin-7.0.0-Power_Macintosh-2.3 >>> -Lbuild/temp.darwin-7.0.0-Power_Macintosh-2.3 >>> -L/usr/local/lib/gcc/powerpc-apple-darwin6.6/3.4 -L/usr/local/lib >>> -ldfftpack -lrfftw -lfftw -lrootfind -lsuperlu -lmyblas -lc_misc >>> -lcephes -lg2c -o >>> build/lib.darwin-7.0.0-Power_Macintosh-2.3/scipy/fftpack/_fftpack.so >>> ld: Undefined symbols: >>> restFP >>> saveFP >>> error: command 'gcc' failed with exit status 1 >>> >>> I have tried using the -faltivec flag, but with the same result. >>> >>> On Nov 25, 2003, at 3:04 AM, Pearu Peterson wrote: >>> >>>> >>>> >>>> On Mon, 24 Nov 2003, Travis E. Oliphant wrote: >>>> >>>>> You can also just not compile sparse support for now. Edit the >>>>> setup.py >>>>> file and comment out the line which includes the sparse package. >>>> >>>> With the recent CVS one can disable Scipy modules by adding their >>>> names to ignore_packages list at the end of setup.py file. >>>> Or create a customized setup file similar to setup_no_chaco.py. >>>> For example, create setup_osx.py file containing: >>>> >>>> from setup import setup_package >>>> ignore_packages = ['chaco','kiva','freetype','traits', 'sparse'] >>>> if __name__ == "__main__": >>>> setup_package(ignore_packages) >>>> >>>> and then use it instead of setup.py. >>>> >>>> Pearu >>>> >>>> _______________________________________________ >>>> SciPy-user mailing list >>>> SciPy-user at scipy.net >>>> http://www.scipy.net/mailman/listinfo/scipy-user >>>> >>>> >>> -- >>> 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 >> _______________________________________________ >> SciPy-user mailing list >> SciPy-user at scipy.net >> http://www.scipy.net/mailman/listinfo/scipy-user >> > -- > 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 -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 2357 bytes Desc: not available URL: From managan at llnl.gov Tue Nov 25 11:31:08 2003 From: managan at llnl.gov (Rob Managan) Date: Tue, 25 Nov 2003 08:31:08 -0800 Subject: [SciPy-user] more OSX build problems In-Reply-To: References: Message-ID: >On Tue, 25 Nov 2003, Christopher Fonnesbeck wrote: > >> Thanks Pearu, it seems to get beyond that now, but choking on some >> undefined symbols further along: >> >> gcc -Wl,-F. -Wl,-F. -bundle -framework Python -faltivec > > > >> ld: Undefined symbols: >> restFP >> saveFP >> error: command 'gcc' failed with exit status 1 > >Hmm, these symbols should be OSX related. Anyone having ideas how to fix >it? Try -dynamiclib option (that I pick up from google search). > >Pearu I found this note on the unix-porting list that apple runs At 10:01 PM -0800 11/13/03, unix-porting-request at lists.apple.com wrote: >List-Subscribe: >, > > >At 11/29/02 9:26 AM, Yves de Champlain wrote: > > > > I get these undefined symbols quite often with 10.2 : > > > > saveFP > > restFP > >Link with either /usr/lib/libcc_dynamic.a or /usr/lib/libgcc.a, >depending on >your needs. It's always handy to do: > >%cd /usr/lib >%nm -o lib* | egrep 'T saveFP' > >for example. > >-Conrad -- *-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*- Rob Managan email managan at llnl.gov LLNL phone: 925-423-0903 P.O. Box 808, L-095 FAX: 925-422-3389 Livermore, CA 94551-0808 From cjw at sympatico.ca Tue Nov 25 14:08:30 2003 From: cjw at sympatico.ca (Colin J. Williams) Date: Tue, 25 Nov 2003 14:08:30 -0500 Subject: [SciPy-user] PyMatrix: Announcement Message-ID: <3FC3A8AE.6030204@sympatico.ca> In response to a number of requests, the PyMatrix package is now available for download as a zip file. Colin W. > PyMatrix is available for test and review. > http://www3.sympatico.ca/cjw > > PyMatrix provides access to basic matrix arithmetic, using Python and > numarray. > > Examples: > A * B => the product of matrices A and B > A.I => the inverse of matrix A > A.EVectors => the eigenvectors of A > A.var(0) => the variances of the columns of A > (a.T*a).I * a.T*b => the solution (x) for a * x = b, > where a is a matrix and b a column vector > > This package was developed on a Windows XP. I would appreciate > comments, particularly with respect to usage on other systems. > > Colin W. > From chris at fonnesbeck.org Tue Nov 25 14:09:41 2003 From: chris at fonnesbeck.org (Christopher Fonnesbeck) Date: Tue, 25 Nov 2003 14:09:41 -0500 Subject: [SciPy-user] more OSX build problems In-Reply-To: References: Message-ID: Success! I got scipy to build on OSX, finally, using the HPC g77 binary with gcc3.3. I needed to add a "-lcc_dynamic" flag for gcc. Thanks to Guarav Khanna for the tip. Now to test. C. On Nov 25, 2003, at 11:31 AM, Rob Managan wrote: >> On Tue, 25 Nov 2003, Christopher Fonnesbeck wrote: >> >>> Thanks Pearu, it seems to get beyond that now, but choking on some >>> undefined symbols further along: >>> >>> gcc -Wl,-F. -Wl,-F. -bundle -framework Python -faltivec >> >> >> >>> ld: Undefined symbols: >>> restFP >>> saveFP >>> error: command 'gcc' failed with exit status 1 >> >> Hmm, these symbols should be OSX related. Anyone having ideas how to >> fix >> it? Try -dynamiclib option (that I pick up from google search). >> >> Pearu > > I found this note on the unix-porting list that apple runs > > At 10:01 PM -0800 11/13/03, unix-porting-request at lists.apple.com wrote: >> List-Subscribe: >> , >> >> >> At 11/29/02 9:26 AM, Yves de Champlain wrote: >> > >> > I get these undefined symbols quite often with 10.2 : >> > >> > saveFP >> > restFP >> >> Link with either /usr/lib/libcc_dynamic.a or /usr/lib/libgcc.a, >> depending on >> your needs. It's always handy to do: >> >> %cd /usr/lib >> %nm -o lib* | egrep 'T saveFP' >> >> for example. >> >> -Conrad > > -- > *-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*- > Rob Managan email managan at llnl.gov > LLNL phone: 925-423-0903 > P.O. Box 808, L-095 FAX: 925-422-3389 > Livermore, CA 94551-0808 > > _______________________________________________ > SciPy-user mailing list > SciPy-user at scipy.net > http://www.scipy.net/mailman/listinfo/scipy-user > > -- 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 Wed Nov 26 10:53:51 2003 From: chris at fonnesbeck.org (Christopher Fonnesbeck) Date: Wed, 26 Nov 2003 10:53:51 -0500 Subject: [SciPy-user] PyMC 0.2 released Message-ID: I'd like to announce the first public release of PyMC, a Markovian system analysis package that includes modules for Markov chain Monte Carlo (MCMC), Kalman filtering (linear and non-linear), reinforcement learning, and function approximation. It relies on (among other things) scipy, mainly for plotting at this point (I wanted to use scipy for its probability distributions as well, but these were quite slow -- maybe in the future). Its been tested on OSX, Linux and Windows. Its still a bit rough around the edges, but I've used it for a wide array of applications already, mostly associated with natural resource management problems. For more information and downloads: http://pymc.sourceforge.net 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 gvermeul at grenoble.cnrs.fr Wed Nov 26 14:59:24 2003 From: gvermeul at grenoble.cnrs.fr (Gerard Vermeulen) Date: Wed, 26 Nov 2003 20:59:24 +0100 Subject: [SciPy-user] ANN: PyQwt-3.8 Message-ID: <20031126205924.62dc5947.gvermeul@grenoble.cnrs.fr> PyQwt-3.8 = FAST and EASY data plotting for Python and (Py)Qt. PyQwt is a set of Python bindings for the Qwt C++ class library which extends the Qt framework with widgets for scientific and engineering applications. It supports the use of PyQt, Qt, Qwt, the Numerical Python extensions (either Numeric, or numarray or both) and optionally SciPy in a GUI Python application or in an interactive Python session. The home page of PyQwt is http://pyqwt.sourceforge.net. Main changes in PyQwt-3.8: 1. PyQwt inherits from Qwt the new classes QwtPicker, QwtPlotPicker and QwtPlotZoomer. QwtPicker can make any widget react to events by event filtering. QwtPlotPicker is a QwtPicker taylored to the plot canvas. QwtPlotZoomer implements infinite deep zooming on the plot canvas. 2. PyQwt inherits from Qwt the support for Qt's rich text format in QwtPlot to display equations like E=mc2. Caveat: in principle, a truetype font with full unicode support is better to typeset equations. 3. Bug fixes to support event filtering code (see EventFilterDemo.py). 4. Conversion of a QImage to a Numeric array or numarray array. 5. Better support for cloning a plot into Grace from the command line interpreter friendly plot classes. 6. PyQwt-3.8 works with PyQt-3.8.1 downto -3.6 and Qt-3.2.3 downto -2.3.0. 7. License change: GPL with permission for dynamic linking to commercial, educational and non-commercial versions of Qt, PyQt and sip. In this case PyQwt becomes a free plug-in for a non-free program. Gerard Vermeulen From joe at enthought.com Wed Nov 26 17:34:08 2003 From: joe at enthought.com (Joe Cooper) Date: Wed, 26 Nov 2003 16:34:08 -0600 Subject: [SciPy-user] Testing SpamAssassin installation Message-ID: <3FC52A76.9090204@enthought.com> Hi all, I've just installed SpamAssassin, to work alongside SpamBayes on the SciPy mailing lists. Though very few spams are getting through these days, because we don't permit posting by non-subscribers, it is still a burden to sift through the moderated requests to let the 5% legitimate messages through. Anyway, feel free to ignore this message, as it it just a test to be sure we're still delivering. Thanks. From joe at enthought.com Wed Nov 26 17:52:53 2003 From: joe at enthought.com (Joe Cooper) Date: Wed, 26 Nov 2003 16:52:53 -0600 Subject: [SciPy-user] Testing again Message-ID: <3FC52EDB.9070601@enthought.com> Oops...It helps if I uncomment the procmail rule that makes it all work before testing. Just testing the new SpamAssassin installation again. From wagner.nils at vdi.de Thu Nov 27 08:37:05 2003 From: wagner.nils at vdi.de (Nils Wagner) Date: Thu, 27 Nov 2003 14:37:05 +0100 Subject: [SciPy-user] MatrixMarket I/O Functions for scipy Message-ID: <20031127133709.183003EB25@www.scipy.com> Dear experts, Has anyone written a module for importing matrices which are stored in MatrixMarket or Harwell-Boeing format ? Afaik, there exists only a Fortran routine for this purpose. http://math.nist.gov/~KRemington/mmio/mmiof77.html It would be great if these formats will be supported in scipy as well. Any comment or suggestion ? Nils From nwagner at mecha.uni-stuttgart.de Thu Nov 27 10:08:02 2003 From: nwagner at mecha.uni-stuttgart.de (Nils Wagner) Date: Thu, 27 Nov 2003 16:08:02 +0100 Subject: [SciPy-user] eps, jpg and png within scipy Message-ID: <3FC61352.F0B87DF@mecha.uni-stuttgart.de> Hi all, I am using pdflatex very frequently. AFAIK, \includegraphics is restricted to jpg and png format. Hence, eps-file cannot be used within pdflatex. Does scipy xplt support these formats (*.jpg, *.png) ? Nils From pearu at scipy.org Thu Nov 27 10:13:50 2003 From: pearu at scipy.org (Pearu Peterson) Date: Thu, 27 Nov 2003 09:13:50 -0600 (CST) Subject: [SciPy-user] eps, jpg and png within scipy In-Reply-To: <3FC61352.F0B87DF@mecha.uni-stuttgart.de> Message-ID: On Thu, 27 Nov 2003, Nils Wagner wrote: > Hi all, > > I am using pdflatex very frequently. > AFAIK, \includegraphics is restricted > to jpg and png format. Hence, > eps-file cannot be used within pdflatex. You can use ps2pdf to convert eps-file to pdf-file and then \includegraphics this pdf-file to a pdflatex document. HTH, Pearu From oliphant at ee.byu.edu Thu Nov 27 15:10:05 2003 From: oliphant at ee.byu.edu (Travis Oliphant) Date: Thu, 27 Nov 2003 13:10:05 -0700 Subject: [SciPy-user] eps, jpg and png within scipy In-Reply-To: <3FC61352.F0B87DF@mecha.uni-stuttgart.de> References: <3FC61352.F0B87DF@mecha.uni-stuttgart.de> Message-ID: <3FC65A1D.70300@ee.byu.edu> Nils Wagner wrote: > Hi all, > > I am using pdflatex very frequently. > AFAIK, \includegraphics is restricted > to jpg and png format. Hence, > eps-file cannot be used within pdflatex. > > Does scipy xplt support these formats (*.jpg, *.png) ? > A couple of things. You should try using LyX (it can use pdflatex to output pdf files and does image conversions automatically). Then, try at xplt.eps(, pdf=1) -Travis > Nils > > _______________________________________________ > SciPy-user mailing list > SciPy-user at scipy.net > http://www.scipy.net/mailman/listinfo/scipy-user From oliphant at ee.byu.edu Thu Nov 27 16:35:47 2003 From: oliphant at ee.byu.edu (Travis Oliphant) Date: Thu, 27 Nov 2003 14:35:47 -0700 Subject: [SciPy-user] Segfaults with weave Message-ID: <3FC66E33.3030008@ee.byu.edu> With Python 2.3.2 on Mandrake 9.2 I ran into trouble with segfaults and weave. I traced the problem to bsddb module in Python 2.3.2. Apparently it segfaults under my configuration when trying to open a file. If you are running Python 2.3.2 could you try import bsddb bsddb.hashopen('anyfile','c') and see if that works for you. I fixed the problem by remmoving _bsddb.so (so that anydbm doesn't use the bsddb module as a db). Now, weave works fine. Could somebody else veryify to see if bsddb is working on Python 2.3.2 for you? -Travis Oliphant From karshi.hasanov at utoronto.ca Thu Nov 27 16:59:58 2003 From: karshi.hasanov at utoronto.ca (Karshi F.Hasanov) Date: Thu, 27 Nov 2003 16:59:58 -0500 Subject: [SciPy-user] source_build_help Message-ID: <200311271659.58913.karshi.hasanov@utoronto.ca> Hi all, I am running Gentoo Linux, and having problems installing SciPy from the source: ------------------------------------------- found GNU Fortran (GCC 3.2.3 20030422 (Gentoo Linux 1.4 3.2.3-r2, propolice)) 3.2.3 20030422 (Gentoo Linux 1.4 3.2.3-r2), propolice using Gnu version='3.2.3' F77='g77' F77FLAGS=' -Wall -fno-second-underscore -fPIC ' F77OPT=' -O3 -funroll-loops -mmmx -msse2 -malign-double -fomit-frame-pointer ' LIBS='-lg2c' Fortran compiler building 'statlib' library ar -cur build/temp.linux-i686-2.2/libstatlib.a build/temp.linux-i686-2.2/ spearman.o build/temp.linux-i686-2.2/ansari.o build/temp.l-i686-2.2/swilk.o ranlib build/temp.linux-i686-2.2/libstatlib.a building 'fitpack' library g77 -Wall -fno-second-underscore -fPIC -O3 -funroll-loops -mmmx -msse2 -malign-double -fomit-frame-pointer -c Lib/interpo/fitpack/pogrid.f -o build/temp.linux-i686-2.2/pogrid.o /tmp/ccFREPBy.s: Assembler messages: /tmp/ccFREPBy.s:215: Error: suffix or operands invalid for `movd' error: failure during compile (exit status = 256) ---------------------------------------------------------- Any suggestions? From pearu at scipy.org Fri Nov 28 04:55:39 2003 From: pearu at scipy.org (Pearu Peterson) Date: Fri, 28 Nov 2003 03:55:39 -0600 (CST) Subject: [SciPy-user] source_build_help In-Reply-To: <200311271659.58913.karshi.hasanov@utoronto.ca> Message-ID: On Thu, 27 Nov 2003, Karshi F.Hasanov wrote: > Hi all, > > I am running Gentoo Linux, and having problems installing > SciPy from the source: > ------------------------------------------- > found GNU Fortran (GCC 3.2.3 20030422 (Gentoo Linux 1.4 3.2.3-r2, propolice)) > 3.2.3 20030422 (Gentoo Linux 1.4 3.2.3-r2), propolice > using Gnu > version='3.2.3' > F77='g77' > F77FLAGS=' -Wall -fno-second-underscore -fPIC ' > F77OPT=' -O3 -funroll-loops -mmmx -msse2 -malign-double > -fomit-frame-pointer ' > LIBS='-lg2c' Fortran compiler > building 'statlib' library > ar -cur build/temp.linux-i686-2.2/libstatlib.a build/temp.linux-i686-2.2/ > spearman.o build/temp.linux-i686-2.2/ansari.o build/temp.l-i686-2.2/swilk.o > ranlib build/temp.linux-i686-2.2/libstatlib.a > building 'fitpack' library > g77 -Wall -fno-second-underscore -fPIC -O3 -funroll-loops -mmmx -msse2 > -malign-double -fomit-frame-pointer -c Lib/interpo/fitpack/pogrid.f -o > build/temp.linux-i686-2.2/pogrid.o > /tmp/ccFREPBy.s: Assembler messages: > /tmp/ccFREPBy.s:215: Error: suffix or operands invalid for `movd' > error: failure during compile (exit status = 256) > ---------------------------------------------------------- > Any suggestions? This issue has been dealt in http://scipy.net/pipermail/scipy-user/2003-March/001359.html but the supporting code is not implemented in scipy_distutils. To do that, some more information is needed: Is this the problem only on Gentoo system with GCC-3.2.3? Can you try another version of GCC? Is scipy_core/scipy_distutils/command/cpuinfo.py detecting sse2 feature correctly on your machine? Check the /proc/cpuinfo file. I tried building scipy using GCC-3.2.3 compiler on P4 Debian sid system and it was succesful. So, it would be unfair to non-Genoo users to disable -sse2 flag for GCC-3.2.x compilers. How to determine that Python runs on Gentoo system? Then we could disable -sse2 only for Gentoo. Pearu From hetal75 at hotmail.com Fri Nov 28 06:33:43 2003 From: hetal75 at hotmail.com (Hetal Patel) Date: Fri, 28 Nov 2003 11:33:43 +0000 Subject: [SciPy-user] Pythonw crashes if I try to plot Complex data points Message-ID: Whenever I try to plot complex data using either plt or gplt from any pyhton shell(Python,PythonWin,IDLE,PyCrust). windows reports that pythonw has crashed. I know that it's a bit unfair to ask the plot routines to plot complex data but I would expect an error, not python to crash. I'm running the latest SciPy release on Windows XP. Example code below crashes. >>>from scipy import * >>>a = arange(1,10,0.2) >>>b = a + 2.j >>>b array([ 1. +2.j, 1.2+2.j, 1.4+2.j, 1.6+2.j, 1.8+2.j, 2. +2.j, 2.2+2.j, 2.4+2.j, 2.6+2.j, 2.8+2.j, 3. +2.j, 3.2+2.j, 3.4+2.j, 3.6+2.j, 3.8+2.j, 4. +2.j, 4.2+2.j, 4.4+2.j, 4.6+2.j, 4.8+2.j, 5. +2.j, 5.2+2.j, 5.4+2.j, 5.6+2.j, 5.8+2.j, 6. +2.j, 6.2+2.j, 6.4+2.j, 6.6+2.j, 6.8+2.j, 7. +2.j, 7.2+2.j, 7.4+2.j, 7.6+2.j, 7.8+2.j, 8. +2.j, 8.2+2.j, 8.4+2.j, 8.6+2.j, 8.8+2.j, 9. +2.j, 9.2+2.j, 9.4+2.j, 9.6+2.j, 9.8+2.j]) >>>gplt.plot(b) At this point python\pythonw crashes with the following error "python.exe has encountered a problem and needs to close. We are sorry for the inconvenience." If anyone can help me resolve this then I would greatly appreciate it. _________________________________________________________________ Use MSN Messenger to send music and pics to your friends http://www.msn.co.uk/messenger From nwagner at mecha.uni-stuttgart.de Fri Nov 28 08:18:34 2003 From: nwagner at mecha.uni-stuttgart.de (Nils Wagner) Date: Fri, 28 Nov 2003 14:18:34 +0100 Subject: [SciPy-user] Harwell-Boeing format and xreadhb in SuperLU ? Message-ID: <3FC74B2A.15A40D74@mecha.uni-stuttgart.de> Hi all, I am wondering if one can import matrices stored in Harwell-Boeing format with scipy. I found some ANSI C files in scipy/Lib/sparse/SuperLU2.0/SRC (xreadhb.c) If this function is alredy available, how can I import the Chuck matrix ck656.mtx ? A small example in scipy would be appreciated. Thanks in advance Nils From jdhunter at ace.bsd.uchicago.edu Fri Nov 28 10:01:56 2003 From: jdhunter at ace.bsd.uchicago.edu (John Hunter) Date: Fri, 28 Nov 2003 09:01:56 -0600 Subject: [SciPy-user] eps, jpg and png within scipy In-Reply-To: (Pearu Peterson's message of "Thu, 27 Nov 2003 09:13:50 -0600 (CST)") References: Message-ID: >>>>> "Pearu" == Pearu Peterson writes: Pearu> You can use ps2pdf to convert eps-file to pdf-file and then Pearu> \includegraphics this pdf-file to a pdflatex document. This is a very good way to go, but it is important that you do a couple of things to get the fonts to work so the pdf's don't look blurry when viewed with a pdf viewer, eg, acroread. You must have a recent version of gv (7.x) and then use the following flags when making the psd file > dvips -Pcmz -Pamz -G1 -Ppdf -o final3.ps final3.dvi > ps2pdf final3.ps This works well, and you can include ps or eps directly into your LaTeX documents. JDH From nwagner at mecha.uni-stuttgart.de Fri Nov 28 10:17:11 2003 From: nwagner at mecha.uni-stuttgart.de (Nils Wagner) Date: Fri, 28 Nov 2003 16:17:11 +0100 Subject: [SciPy-user] eps, jpg and png within scipy References: Message-ID: <3FC766F7.2ACF0DA4@mecha.uni-stuttgart.de> John Hunter schrieb: > > >>>>> "Pearu" == Pearu Peterson writes: > > Pearu> You can use ps2pdf to convert eps-file to pdf-file and then > Pearu> \includegraphics this pdf-file to a pdflatex document. > > This is a very good way to go, but it is important that you do a > couple of things to get the fonts to work so the pdf's don't look > blurry when viewed with a pdf viewer, eg, acroread. > > You must have a recent version of gv (7.x) and then use the following > flags when making the psd file > > > dvips -Pcmz -Pamz -G1 -Ppdf -o final3.ps final3.dvi > > ps2pdf final3.ps > > This works well, and you can include ps or eps directly into your > LaTeX documents. > > JDH > gv (7.x) Is that a typo ? http://wino.physik.uni-mainz.de/~plass/gv/ I found gv 3.5.8 Nils > _______________________________________________ > SciPy-user mailing list > SciPy-user at scipy.net > http://www.scipy.net/mailman/listinfo/scipy-user From pearu at scipy.org Fri Nov 28 11:09:59 2003 From: pearu at scipy.org (Pearu Peterson) Date: Fri, 28 Nov 2003 10:09:59 -0600 (CST) Subject: [SciPy-user] eps, jpg and png within scipy In-Reply-To: Message-ID: On Fri, 28 Nov 2003, John Hunter wrote: > >>>>> "Pearu" == Pearu Peterson writes: > > Pearu> You can use ps2pdf to convert eps-file to pdf-file and then > Pearu> \includegraphics this pdf-file to a pdflatex document. > > This is a very good way to go, but it is important that you do a > couple of things to get the fonts to work so the pdf's don't look > blurry when viewed with a pdf viewer, eg, acroread. > > You must have a recent version of gv (7.x) and then use the following > flags when making the psd file > > > dvips -Pcmz -Pamz -G1 -Ppdf -o final3.ps final3.dvi > > ps2pdf final3.ps > > This works well, and you can include ps or eps directly into your > LaTeX documents. I think Nils is using the process latex->pdf, not latex->dvi->ps->pdf that you are describing. But indeed, latex->dvi->ps->pdf requires some consideration in the dvi->ps step. My experience is that dvips -Pwww -o file.ps file.dvi is enough to produce a pdf file with proper fonts included when using ps2pdf. And of course, gs-7.x (not gv) is what one must have in order to ps2pdf get fonts right. Regards, Pearu From chris at fonnesbeck.org Fri Nov 28 22:43:53 2003 From: chris at fonnesbeck.org (Christopher Fonnesbeck) Date: Fri, 28 Nov 2003 22:43:53 -0500 Subject: [SciPy-user] mysterious problem with gplt.png Message-ID: <4057F368-221E-11D8-9F7D-000A956FDAC0@fonnesbeck.org> I now have scipy more-or-less happily functioning on my OSX machine (thanks to all that offered help and advice!). I am, however, running into an odd problem with exporting plots using gplt to a png file. gplt.png and gplt.output both refuse to output to the specified file, even though the methods seem to run without error. I'm not even sure how to debug this. Any ideas? Here is some sample code: from scipy.gplt import * def plot_time_series(data,name,xlab='Time',ylab='State'): 'Plot (mutiple) time series' plot(data) print 'Plotting',name grid('off') xtitle(xlab) ytitle(ylab) png("%s.png" % name) close() Thanks, -- 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 pearu at scipy.org Sat Nov 29 04:38:06 2003 From: pearu at scipy.org (Pearu Peterson) Date: Sat, 29 Nov 2003 03:38:06 -0600 (CST) Subject: [SciPy-user] mysterious problem with gplt.png In-Reply-To: <4057F368-221E-11D8-9F7D-000A956FDAC0@fonnesbeck.org> Message-ID: On Fri, 28 Nov 2003, Christopher Fonnesbeck wrote: > I now have scipy more-or-less happily functioning on my OSX machine > (thanks to all that offered help and advice!). I am, however, running > into an odd problem with exporting plots using gplt to a png file. > gplt.png and gplt.output both refuse to output to the specified file, > even though the methods seem to run without error. I'm not even sure > how to debug this. Any ideas? Here is some sample code: > > from scipy.gplt import * > def plot_time_series(data,name,xlab='Time',ylab='State'): > 'Plot (mutiple) time series' > > plot(data) > print 'Plotting',name > grid('off') > xtitle(xlab) > ytitle(ylab) > png("%s.png" % name) > close() The above function works here fine (png images are outputed to specified files). Do you get any errors? What are they? Just in any case, scipy.gplt requires gnuplot installed. Pearu From chris at fonnesbeck.org Sat Nov 29 12:15:41 2003 From: chris at fonnesbeck.org (Christopher Fonnesbeck) Date: Sat, 29 Nov 2003 12:15:41 -0500 Subject: [SciPy-user] mysterious problem with gplt.png In-Reply-To: References: Message-ID: On Nov 29, 2003, at 4:38 AM, Pearu Peterson wrote: > > > On Fri, 28 Nov 2003, Christopher Fonnesbeck wrote: > >> I now have scipy more-or-less happily functioning on my OSX machine >> (thanks to all that offered help and advice!). I am, however, running >> into an odd problem with exporting plots using gplt to a png file. >> gplt.png and gplt.output both refuse to output to the specified file, >> even though the methods seem to run without error. I'm not even sure >> how to debug this. Any ideas? Here is some sample code: >> >> from scipy.gplt import * >> def plot_time_series(data,name,xlab='Time',ylab='State'): >> 'Plot (mutiple) time series' >> >> plot(data) >> print 'Plotting',name >> grid('off') >> xtitle(xlab) >> ytitle(ylab) >> png("%s.png" % name) >> close() > > The above function works here fine (png images are outputed > to specified files). Do you get any errors? What are they? > Just in any case, scipy.gplt requires gnuplot installed. > > Pearu > > That's part of the trouble -- there are no errors. I have gnuplot installed, and plots are generated without any trouble. However, when I call png, no file is written. 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 chris at fonnesbeck.org Sun Nov 30 15:16:05 2003 From: chris at fonnesbeck.org (Christopher Fonnesbeck) Date: Sun, 30 Nov 2003 15:16:05 -0500 Subject: [SciPy-user] more OSX build problems In-Reply-To: References: Message-ID: <06C06AB5-2372-11D8-9C3B-000A956FDAC0@fonnesbeck.org> On Nov 25, 2003, at 4:35 PM, Rob Managan wrote: > At 2:38 PM -0500 11/25/03, Christopher Fonnesbeck wrote: >> X-Spam-Status: No, hits=0.0 required=5.0 >> From: Christopher Fonnesbeck >> Subject: Re: [SciPy-user] more OSX build problems >> Date: Tue, 25 Nov 2003 14:38:21 -0500 >> To: Rob Managan >> >> The g77 3.4 binary for OSX is available at the HPC site >> (http://hpc.sourceforge.net), and it works with gcc 3.3 that comes >> with panther (or in the latest developer tools). >> >> You need to specify the CFLAGS as follows: >> >> CFLAGS="-faltivec -lcc_dynamic" >> >> Then build and install the cvs version using "setup_no_chaco.py" . >> >> I *think* thats 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 > > I guess that means you installed the prereqs like ATLAS, LAPACK, f2py, > fftw. > Any major pointers on those or are they straightforward? > > I know I tried ATLAS once and while it worked it was not clear it was > optimal. > > I am going to give it a shot with the f77 based on v3.1 since I am > still using 10.2.8 and gcc v3.1 The only one you will have to install explicitly is f2py; ATLAS, LAPACK and fftw are available when you install the developer tools. I've been using scipy on Panther for a few days with few problems. The only major one is the inability to output graphics from gplt (see my previous posting on this). 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