pbrain-users Mailing List for brain analysis software in python

SourceForge Headquarters 1320 Columbia Street Suite 310 San Diego, CA 92101 +1 (858) 422-6466

Hey. I'd like to have a look at Loc3D, could you send me the link
to your debian repository?

Thanks

Martin

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi John,

if you have time, I hope you can shed some light on my questions to the
nipy-dev list below.

Thanks, Darren

- -------- Original Message --------
Subject: [Nipy-devel] data structures for brain surfaces
Date: Tue, 06 Mar 2007 16:56:29 -0800
From: Darren Weber <dw...@ra...>
Reply-To: Dar...@ra...
Organization: UCSF Department of Radiology
To: nipy-devel <nip...@ne...>

Consider some simple geometric data files, eg:

http://www.geomview.org/docs/oogltour.html

how would you best work with them in nipy/pbrain?  Would you read the
data into a numpy array or matrix?  Would you use OpenGL data structures
or maybe VTK data structures?

I assume that however this is done, the same process would apply for
data IO and visualization for surfaces derived from MRI segmentation and
tesellation (eg, freesurfer data, among others).

I guess my preference is to try to stay close to any c/c++ wrapping.  If
OpenGL, vtk, etc. are wrapped somehow (swig), I guess it's best to use
their functions directly.  However, the data structures involved are not
trivial and they are not necessarily portable from one package to another.

I'm wondering about how pyOpenGL and pyVTK relate to the numpy
developments.  maybe there is no dependency of pyOpenGL or pyVTK on
numpy?  Maybe they do depend on numpy, are they catching up with the new
numpy?

Take care, Darren

_______________________________________________
Nipy-devel mailing list
Nip...@ne...
http://lists.neuroimaging.scipy.org/mailman/listinfo/nipy-devel

- --

Darren L. Weber, Ph.D.
Postdoctoral Scholar

Dynamic Neuroimaging Laboratory,
UCSF Department of Radiology,
185 Berry Street, Suite 350, Box 0946,
San Francisco, CA 94107, USA.

Tel: +1 415 353-9444
Fax: +1 415 353-9421
www: http://dnl.ucsf.edu/users/dweber
pgp: http://dnl.ucsf.edu/users/dweber/dweber_pgp.html

"To explicate the uses of the brain seems as difficult
a task as to paint the soul, of which it is commonly
said, that it understands all things but itself."
  Thomas Willis (The Anatomy of the Brain and Nerves, 1664)

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.3 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFF8e7JxaCYN7qs0v4RApS2AJ9I98MktLjIW4Q6ncSNt13p7afUnACdHYpx
zL5iIZXEiH7wuTNZXSlCYSs=
=kxRp
-----END PGP SIGNATURE-----

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi John,

I'm wondering if Pbrain is still in development or is it now being
integrated into NiPy?

Best, Darren

- --

Darren L. Weber, Ph.D.
Postdoctoral Scholar

Dynamic Neuroimaging Laboratory,
UCSF Department of Radiology,
185 Berry Street, Suite 350, Box 0946,
San Francisco, CA 94107, USA.

Tel: +1 415 353-9444
Fax: +1 415 353-9421
www: http://dnl.ucsf.edu/users/dweber

"To explicate the uses of the brain seems as difficult
a task as to paint the soul, of which it is commonly
said, that it understands all things but itself."
  Thomas Willis (The Anatomy of the Brain and Nerves, 1664)
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.3 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFFdxqpxaCYN7qs0v4RAoFfAKCmAH2K9XwgsUMzmDX7G13C384RhQCgiM9i
cXFWriyn6DOCsDU5/9TYhyA=
=igot
-----END PGP SIGNATURE-----

 I'm running Fedora Core 5.  I would appreciate a pointer to your Debian repository, as it appears that I have most of the packages identified in your README doc already installed.  Thanks in advance.  

_______________________________________________
Join Excite! - http://www.excite.com
The most personalized portal on the Web!

Hi,
I am interested in Pbrain.
I would like a pointer to the debian repository in order to get additional information
or download Pbrain.

Thank you very much.

Ruediger

______________________________________________________________________
XXL-Speicher, PC-Virenschutz, Spartarife & mehr: Nur im WEB.DE Club!		
Jetzt gratis testen! http://freemail.web.de/home/landingpad/?mc=021130

Is there a Debian Sarge package for pbrain?  This is what I see:

dnlweber:/data/papers/ucsf_hin# apt-cache search pbrain
egroupware-phpbrain - eGroupWare phpbrain application
phpgroupware-phpbrain - phpGroupWare phpbrain module

using stable repositories.  Do you have a site which can be added to the 
sources list?

Thanks, Darren

-- 

Darren L. Weber, Ph.D.
Visiting Postdoctoral Scholar

Dynamic Neuroimaging Laboratory, Department of Radiology,
University of California, San Francisco,
185 Berry Street, Suite 350, Box 0946,
San Francisco, CA 94107, USA.

Tel: +1 415 353-9444
Fax: +1 415 353-9421
www: http://dnl.ucsf.edu/users/dweber

"To explicate the uses of the brain seems as difficult
a task as to paint the soul, of which it is commonly
said, that it understands all things but itself."
--Thomas Willis (The Anatomy of the Brain and Nerves, 1664)

>>>>> "Morgan" == Morgan Hough <mh...@fm...> writes:

    Morgan> I haven't had time to revisit pbrain issues on fc4 and Mac
    Morgan> OS X recently but I tried the latest subversion revisions
    Morgan> of numpy, scipy and pbrain last night and I am happy to
    Morgan> report that everything seems to be working fine. Other
    Morgan> than numpy and scipy, there are only two dependencies that
    Morgan> aren't in Fedora Extras.

    Morgan> A couple quick questions.

    Morgan> When you get a new Neuroscan .cnt file (I am right in
    Morgan> thinking you use Neuroscan right?), are you using other
    Morgan> software to do any preprocessing or move it into the
    Morgan> database?

    Morgan> I see from the preferences that there is a zope pane. What
    Morgan> are you using zope for? Is this for displaying subject
    Morgan> data and results via the web?  That would be very cool.

Yes, there is a zope mysql interface that stores data and metadata
about the EEG, associated files, grids, etc, and eegview is aware of
this and can pull data from the web with local filesystem caching.

The basic setup uses mysql to serve the metadata and zope to serve the
eeg data.

    Morgan> I was wondering if you think one could use loc3djr for
    Morgan> determining electrode positions from MRI-compat EEG caps
    Morgan> on the basis of the MRI data?

Certainly seems like this should be a possibility, as long as the data
shows up in the MRI and is of sufficient contrast.  The segmentation
algorithms that loc3djr uses are not terribly sophisticated, and so
they work best on high contrast data like CT.

JDH

I haven't had time to revisit pbrain issues on fc4 and Mac OS X recently
but I tried the latest subversion revisions of numpy, scipy and pbrain
last night and I am happy to report that everything seems to be working
fine. Other than numpy and scipy, there are only two dependencies that
aren't in Fedora Extras.

A couple quick questions.

When you get a new Neuroscan .cnt file (I am right in thinking you use
Neuroscan right?), are you using other software to do any preprocessing
or move it into the database? 

I see from the preferences that there is a zope pane. What are you using
zope for? Is this for displaying subject data and results via the web?
That would be very cool. 

I was wondering if you think one could use loc3djr for determining
electrode positions from MRI-compat EEG caps on the basis of the MRI
data?  

Thanks in advance

Cheers,

-Morgan

-- 
Morgan G. Hough, DPhil Candidate
University of Oxford, Centre for Functional MRI of the Brain (FMRIB)
John Radcliffe Hospital, Headington, Oxford OX3 9DU, United Kingdom
Tel:    +44 (0) 1865 222545  / Fax:  +44 (0) 1865 222717
e-mail: mh...@fm...  / URL:  http://www.fmrib.ox.ac.uk/~mhough

I have worked my way through the dependency issues on Mac OS X and
Fedora Core 4 and I am trying to start eegview and loc3djr. I am using
the new scipy on Fedora Core 4. On both platforms I get the same errors.

eegview.py gives me this traceback:

Traceback (most recent call last):  File
"/sw/lib/python2.4/site-packages/eegview/eegview.py", line 47, in ?
from mpl_windows import ChannelWin, AcorrWin, HistogramWin, SpecWin
File "/sw/lib/python2.4/site-packages/eegview/mpl_windows.py", line 570,
in ?    class SpecWin(MPLWin):  File
"/sw/lib/python2.4/site-packages/eegview/mpl_windows.py", line 571, in
SpecWin    propdlg = SpecProps()  File
"/sw/lib/python2.4/site-packages/eegview/dialogs.py", line 1069, in
__init__    xoptions=gtk.FALSE, yoptions=gtk.FALSE)TypeError: flag
values must be strings, ints or tuplesUpdated RC
file /Users/mhough/.eegviewrc

which looking at .eegviewrc also made me wonder about what mysql setup
is required.

loc3djr gives me this traceback:

Traceback (most recent call last):  File
"/sw/lib/python2.4/site-packages/loc3djr/loc3djr.py", line 5, in ?
from plane_widgets import PlaneWidgetsWithObservers  File
"/sw/lib/python2.4/site-packages/loc3djr/plane_widgets.py", line 24,
in ?    from surf_renderer import SurfRendererProps, SurfRenderWindow
File "/sw/lib/python2.4/site-packages/loc3djr/surf_renderer.py", line
30, in ?    class DecimateFilter(vtk.vtkDecimate):AttributeError:
'module' object has no attribute 'vtkDecimate'

which I think is due to the version of VTK I am using (5.0-0.20051008).
Can I add patented to recent CVS versions? Do I grab patented from an
older version and mix them or do I have to use an older version.

Cheers,

-Morgan

-- 
Morgan G. Hough, DPhil Candidate
University of Oxford, Centre for Functional MRI of the Brain (FMRIB)
John Radcliffe Hospital, Headington, Oxford OX3 9DU, United Kingdom
Tel:    +44 (0) 1865 222545  / Fax:  +44 (0) 1865 222717
e-mail: mh...@fm...  / URL:  http://www.fmrib.ox.ac.uk/~mhough

Hi Darren,

Darren Weber wrote:
> Hi,
> 
> I am new to installation of ATLAS and LAPACK and the integration between
> them is a bit confusing.  I don't often compile and test libraries like
> this, I just want to get scientific python and the R project working
> properly with these libraries on a linux system.  Sorry if these 
> questions are too simple or directed to the wrong email lists.  I have 
> tried to read the documentation for installing ATLAS and LAPACK and also 
> checked documents about this at websites related to scientific python 
> and R.  I have not resolved my questions entirely and I wonder if anyone 
> can help?  The atlas/lapack developers are not easily accessible and 
> there is no user group for newbie support.

here you can find rpms made from the full ATLAS binaries provided at the scipy 
site:

http://amath.colorado.edu/faculty/fperez/tmp/

These are for Pentium3 and Pentium4, and have been used successfully for about 
a year on both architectures, Fedora3.  Feel free to use them.

They include a full lapack, and they go to the location the scipy build script 
expects them to live in, so they should work 'out of the box'.

Regards,

f

Hi,

I am new to installation of ATLAS and LAPACK and the integration between
them is a bit confusing.  I don't often compile and test libraries like
this, I just want to get scientific python and the R project working
properly with these libraries on a linux system.  Sorry if these 
questions are too simple or directed to the wrong email lists.  I have 
tried to read the documentation for installing ATLAS and LAPACK and also 
checked documents about this at websites related to scientific python 
and R.  I have not resolved my questions entirely and I wonder if anyone 
can help?  The atlas/lapack developers are not easily accessible and 
there is no user group for newbie support.

I have a successful build and test for ATLAS.  I have done so under

/usr/src/ATLAS (using a Linux_P4SSE2 platform)

First, I am wondering where to install the libraries, as the make
install does not seem to put them into a default location.  I notice in
setup.py of the numeric python module that it suggests they could be
located in
/usr/lib/atlas
/usr/include/atlas
while another python web page, related to scipy, recommends
/usr/local/lib/atlas
/usr/local/include/atlas

I cannot find any documentation in ATLAS about where to install the
libraries.  I am happy to build and test them under /usr/src/ATLAS, but
I don't want to install this complete build path into /usr/lib etc.

The ATLAS documents recommend creating a combined lapack by running
lapack and then integrating the two lapack libraries from lapack and
atlas.  The lapack documents indicate building lapack with an optimised
blas.  So, I have built ATLAS and now we have the following files:

[root@dnlweber LAPACK]# ls /usr/src/ATLAS/lib/Linux_P4SSE2/
libatlas.a  libcblas.a  libf77blas.a  liblapack.a  libtstatlas.a
Makefile  Make.inc

In the lapack make.inc.LINUX file, there is the following definitions:

#  The location of the libraries to which you will link.  (The
#  machine-specific, optimized BLAS library should be used whenever
#  possible.)
#
BLASLIB      = ../../blas$(PLAT).a
LAPACKLIB    = lapack$(PLAT).a
TMGLIB       = tmglib$(PLAT).a
EIGSRCLIB    = eigsrc$(PLAT).a
LINSRCLIB    = linsrc$(PLAT).a

Now, before I build lapack, what do I put into BLASLIB?   I think there
are several options above:
libatlas.a  libcblas.a  libf77blas.a
but the most likely option is the libf77blas.a, is that right?  In any
case, should I set

BLASLIB      = /usr/src/ATLAS/lib/Linux_P4SSE2/libf77blas.a

or copy this library somewhere and then point it?

Assuming that this works, I understand that I then have to combine the
lapack_LINUX.a from this lapack build with the previous optimized ATLAS
liblapack.a, according to this web page:
http://math-atlas.sourceforge.net/errata.html#completelp

Is that correct?  If I do that, then (again) where do I actually install
the final results of all this on a linux system, into a standard
location so that it will be consistent with a linux LD config.  Should I
setup any environment variables to facilitate linking to the libraries?
I wish there were a step-by-step HOW-TO on this topic for linux that
could standardize this installation that seems to be essential for so
many scientific applications on linux.

Take care, Darren

-- 

Darren L. Weber, Ph.D.
Postdoctoral Scholar

Dynamic Neuroimaging Laboratory, Department of Radiology,
University of California, San Francisco,
185 Berry Street, Suite 350, Box 0946,
San Francisco, CA 94107, USA.

Tel: +1 415 353-9444
Fax: +1 415 353-9421
www: http://dnl.ucsf.edu/users/dweber

"To explicate the uses of the brain seems as difficult
a task as to paint the soul, of which it is commonly
said, that it understands all things but itself."
--Thomas Willis (1664), The Anatomy of the Brain and Nerves.

John,

Maybe I spoke to soon getting excited about passing test.py. After
making the respective files executable I couldn't get either application
to open from an X11 xterm. I attached the error messages I got. Can you
make anything from them? IS there more information I could produce?

Cheers,

-Morgan

Okay. I love MacEnthon and I want native GTK+ (and I got jhbuild working
but there are gettext autoconf issues to fix all over the place) but
eventually I was back at pygtkglext and I finally have a fink package
working. I laughed, I cried, it was a real moment. I am just moving
everything to python 2.4 and submitting my packages (I have brainstat
ready as well since it only needed minc but Jonathan added this CVS
traits.ui stuff that I will never get accepted yet). I should be able to
get darwinports working now if they can fix their scipy package (they
already took my minc package). 

Fedora Extras RPMs are all done thanks to Intrinsic Spin at
monkeyrpms.net who just happens to also be in the Chicago area. The
python ones still need to go through the submission process but it
shouldn't be too long now. 

Cheers,

-Morgan

On Tue, 2005-10-18 at 08:08 -0500, John Hunter wrote: 
> >>>>> "Morgan" == Morgan Hough <mh...@fm...> writes:
> 
>     Morgan> I hear you John but sometimes it saves me so much
>     Morgan> trouble. Especially with packages that have lots of
>     Morgan> dependencies. Nice to hear you have a powerbook
> 
> It does, which is why everyone loves it at first.  After a bit,
> though, you will find a package that doesn't work and you'll be stuck
> and not know where to go.  Everyone goes through the same cycle.
> 
>     Morgan> though. Did you match versions of dependencies to the
>     Morgan> Windows build? I just tried making a package for
>     Morgan> darwinports and I got the same result. Did you get things
>     Morgan> running with 2.8.x or 2.10+? Have you used garnome or
>     Morgan> jhbuild on Mac OS X?
> 
> No, I didn't try and match.  I just grabbed the latest src tarballs
> linked from http://www.macgimp.org/index.php?topic=gtk in the header
> "The Road to GIMP 2.0: Building GTK2+ on Mac OS X".  This requires
> patience and a fair amount of experience building from src, but it
> worked.  Those files have additional dependencies, so yes, it is labor intensive.
> 
>     Morgan> There is a new GTK+ on Mac OS X project that is working
>     Morgan> off of GNOME CVS HEAD. Maybe it will be possible to get a
>     Morgan> native GUI build working soon.
> 
>     Morgan> http://micke.hallendal.net/archives/2005/10/gtk-macosx.html
> 
>     Morgan> The macpython people have many packages ready (mx-base,
>     Morgan> mysql-python, numeric, matplotlib, PIL, PyOpenGL). I
>     Morgan> haven't seen a native Tkinter MayaVi but it should be
>     Morgan> possible these days. No SciPy package yet. Is 0.4.2
>     Morgan> backwards compatible? I guess I just need to get the
> 
> I definitely recommend and use the mac python stuff.
> 
>     Morgan> GTK/PyGTK stuff running then. Of course now I will have to
>     Morgan> go back to building VTK myself again. What version do you
>     Morgan> run on your Mac? You know patented has been removed?
> 
> Yes, I know.  I recompile VTK myself with patented, for OS X, linux
> and win32.
> 
>     Morgan> Will there ever be an Enthought python for Mac OS X?
> 
> It exists!
> 
> http://download.enthought.com/MacEnthon/
> 
> Good luck!
> 
> JDH

>>>>> "Morgan" == Morgan Hough <mh...@fm...> writes:

    Morgan> I hear you John but sometimes it saves me so much
    Morgan> trouble. Especially with packages that have lots of
    Morgan> dependencies. Nice to hear you have a powerbook

It does, which is why everyone loves it at first.  After a bit,
though, you will find a package that doesn't work and you'll be stuck
and not know where to go.  Everyone goes through the same cycle.

    Morgan> though. Did you match versions of dependencies to the
    Morgan> Windows build? I just tried making a package for
    Morgan> darwinports and I got the same result. Did you get things
    Morgan> running with 2.8.x or 2.10+? Have you used garnome or
    Morgan> jhbuild on Mac OS X?

No, I didn't try and match.  I just grabbed the latest src tarballs
linked from http://www.macgimp.org/index.php?topic=gtk in the header
"The Road to GIMP 2.0: Building GTK2+ on Mac OS X".  This requires
patience and a fair amount of experience building from src, but it
worked.  Those files have additional dependencies, so yes, it is labor intensive.

    Morgan> There is a new GTK+ on Mac OS X project that is working
    Morgan> off of GNOME CVS HEAD. Maybe it will be possible to get a
    Morgan> native GUI build working soon.

    Morgan> http://micke.hallendal.net/archives/2005/10/gtk-macosx.html

    Morgan> The macpython people have many packages ready (mx-base,
    Morgan> mysql-python, numeric, matplotlib, PIL, PyOpenGL). I
    Morgan> haven't seen a native Tkinter MayaVi but it should be
    Morgan> possible these days. No SciPy package yet. Is 0.4.2
    Morgan> backwards compatible? I guess I just need to get the

I definitely recommend and use the mac python stuff.

    Morgan> GTK/PyGTK stuff running then. Of course now I will have to
    Morgan> go back to building VTK myself again. What version do you
    Morgan> run on your Mac? You know patented has been removed?

Yes, I know.  I recompile VTK myself with patented, for OS X, linux
and win32.

    Morgan> Will there ever be an Enthought python for Mac OS X?

It exists!

http://download.enthought.com/MacEnthon/

Good luck!

JDH

I hear you John but sometimes it saves me so much trouble. Especially
with packages that have lots of dependencies. Nice to hear you have a
powerbook though. Did you match versions of dependencies to the Windows
build? I just tried making a package for darwinports and I got the same
result. Did you get things running with 2.8.x or 2.10+? Have you used
garnome or jhbuild on Mac OS X? 

There is a new GTK+ on Mac OS X project that is working off of GNOME CVS
HEAD. Maybe it will be possible to get a native GUI build working soon. 

http://micke.hallendal.net/archives/2005/10/gtk-macosx.html

The macpython people have many packages ready (mx-base, mysql-python,
numeric, matplotlib, PIL, PyOpenGL). I haven't seen a native Tkinter
MayaVi but it should be possible these days. No SciPy package yet. Is
0.4.2 backwards compatible? I guess I just need to get the GTK/PyGTK
stuff running then. Of course now I will have to go back to building VTK
myself again. What version do you run on your Mac? You know patented has
been removed?

Will there ever be an Enthought python for Mac OS X?

Cheers,

-Morgan

On Mon, 2005-10-17 at 08:49 -0500, John Hunter wrote:
> >>>>> "Morgan" == Morgan Hough <mh...@fm...> writes:
> 
>     Morgan> John, I only have one more dependency package to make for
>     Morgan> Mac OS X but I am having a problem with pygtkglext. I was
>     Morgan> wondering if you have seen this error with distutils
>     Morgan> before. I have included the traceback as an
>     Morgan> attachment. Someone on the python irc channel suggested a
>     Morgan> problem with dsextra.py but I have had no luck after
>     Morgan> updating that file. Any thoughts?
> 
> Hey Morgan,
> 
> You will live to regret using fink.  Mark my words :-)
> 
> See this thread -- it looks like a known problem with fink and gnome libs
> 
>   http://sourceforge.net/mailarchive/forum.php?thread_id=8022731&forum_id=3200
> 
> I compiled from src on my powerbook -- not fun, admittedly.
> 
> JDH
> 
>     Morgan> Thanks in advance for your time.
> 
>     Morgan> Cheers,
> 
>     Morgan> -Morgan
> 
>     Morgan> /sw/bin/python2.3 setup.py install running install running
>     Morgan> build running build_py running build_ext running
>     Morgan> install_lib Traceback (most recent call last): File
>     Morgan> "setup.py", line 158, in ?  cmdclass={'install_lib':
>     Morgan> PyGtkGLExtInstallLib, File
>     Morgan> "/sw/src/fink.build/root-python23-2.3.5-113/sw/lib/python2.3/distutils/core.py",
>     Morgan> line 149, in setup File
>     Morgan> "/sw/src/fink.build/root-python23-2.3.5-113/sw/lib/python2.3/distutils/dist.py",
>     Morgan> line 907, in run_commands File
>     Morgan> "/sw/src/fink.build/root-python23-2.3.5-113/sw/lib/python2.3/distutils/dist.py",
>     Morgan> line 927, in run_command File
>     Morgan> "/sw/src/fink.build/root-python23-2.3.5-113/sw/lib/python2.3/distutils/command/install.py",
>     Morgan> line 510, in run File "/sw/lib/python2.3/cmd.py", line
>     Morgan> 333, in run_command
>     
>     Morgan>   File
>     Morgan> "/sw/src/fink.build/root-python23-2.3.5-113/sw/lib/python2.3/distutils/dist.py",
>     Morgan> line 927, in run_command File "setup.py", line 61, in run
>     Morgan> self.add_template_option('VERSION', VERSION) File
>     Morgan> "/sw/lib/python2.3/cmd.py", line 112, in __getattr__ if
>     Morgan> self.intro: AttributeError: add_template_option ###
>     Morgan> execution of /sw/bin/python2.3 failed, exit code 1
>     Morgan> Removing build lock...  /sw/bin/dpkg-lockwait -r
>     Morgan> fink-buildlock-pygtkglext-py23-1.0.1-1 (Reading database
>     Morgan> ... 82324 files and directories currently installed.)
>     Morgan> Removing fink-buildlock-pygtkglext-py23-1.0.1-1 ...
>     Morgan> Failed: phase installing: pygtkglext-py23-1.0.1-1 failed
> 
> 
> -------------------------------------------------------
> This SF.Net email is sponsored by:
> Power Architecture Resource Center: Free content, downloads, discussions,
> and more. http://solutions.newsforge.com/ibmarch.tmpl
> _______________________________________________
> Pbrain-users mailing list
> Pbr...@li...
> https://lists.sourceforge.net/lists/listinfo/pbrain-users

>>>>> "Morgan" == Morgan Hough <mh...@fm...> writes:

    Morgan> John, I only have one more dependency package to make for
    Morgan> Mac OS X but I am having a problem with pygtkglext. I was
    Morgan> wondering if you have seen this error with distutils
    Morgan> before. I have included the traceback as an
    Morgan> attachment. Someone on the python irc channel suggested a
    Morgan> problem with dsextra.py but I have had no luck after
    Morgan> updating that file. Any thoughts?

Hey Morgan,

You will live to regret using fink.  Mark my words :-)

See this thread -- it looks like a known problem with fink and gnome libs

  http://sourceforge.net/mailarchive/forum.php?thread_id=8022731&forum_id=3200

I compiled from src on my powerbook -- not fun, admittedly.

JDH

    Morgan> Thanks in advance for your time.

    Morgan> Cheers,

    Morgan> -Morgan

    Morgan> /sw/bin/python2.3 setup.py install running install running
    Morgan> build running build_py running build_ext running
    Morgan> install_lib Traceback (most recent call last): File
    Morgan> "setup.py", line 158, in ?  cmdclass={'install_lib':
    Morgan> PyGtkGLExtInstallLib, File
    Morgan> "/sw/src/fink.build/root-python23-2.3.5-113/sw/lib/python2.3/distutils/core.py",
    Morgan> line 149, in setup File
    Morgan> "/sw/src/fink.build/root-python23-2.3.5-113/sw/lib/python2.3/distutils/dist.py",
    Morgan> line 907, in run_commands File
    Morgan> "/sw/src/fink.build/root-python23-2.3.5-113/sw/lib/python2.3/distutils/dist.py",
    Morgan> line 927, in run_command File
    Morgan> "/sw/src/fink.build/root-python23-2.3.5-113/sw/lib/python2.3/distutils/command/install.py",
    Morgan> line 510, in run File "/sw/lib/python2.3/cmd.py", line
    Morgan> 333, in run_command

    Morgan>   File
    Morgan> "/sw/src/fink.build/root-python23-2.3.5-113/sw/lib/python2.3/distutils/dist.py",
    Morgan> line 927, in run_command File "setup.py", line 61, in run
    Morgan> self.add_template_option('VERSION', VERSION) File
    Morgan> "/sw/lib/python2.3/cmd.py", line 112, in __getattr__ if
    Morgan> self.intro: AttributeError: add_template_option ###
    Morgan> execution of /sw/bin/python2.3 failed, exit code 1
    Morgan> Removing build lock...  /sw/bin/dpkg-lockwait -r
    Morgan> fink-buildlock-pygtkglext-py23-1.0.1-1 (Reading database
    Morgan> ... 82324 files and directories currently installed.)
    Morgan> Removing fink-buildlock-pygtkglext-py23-1.0.1-1 ...
    Morgan> Failed: phase installing: pygtkglext-py23-1.0.1-1 failed

John,

I only have one more dependency package to make for Mac OS X but I am
having a problem with pygtkglext. I was wondering if you have seen this
error with distutils before. I have included the traceback as an
attachment. Someone on the python irc channel suggested a problem with
dsextra.py but I have had no luck after updating that file. Any
thoughts?

Thanks in advance for your time.

Cheers,

-Morgan

John,

I will try updating the script. If I do, I will certainly post to the
list (which I also just joined:).

I pushed for ubuntu at the lab but fc4 won out. There is a lot more
packaged for debian IMHO but que sera, sera.

My vtk-python gives me:

Python 2.4.1 (#1, May 16 2005, 15:19:29)
[GCC 4.0.0 20050512 (Red Hat 4.0.0-5)] on linux2
Type "help", "copyright", "credits" or "license" for more information.
>>> import vtk
>>> d = vtk.vtkMarchingCubes()
>>> c = vtk.vtkDecimate()
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
AttributeError: 'module' object has no attribute 'vtkDecimate'
>>>

I will ask Fabrice directly and see what he says. 

Cheers,

-Morgan

On Fri, 2005-08-12 at 09:57 -0500, John Hunter wrote:
> >>>>> "Morgan" == Morgan Hough <mh...@fm...> writes:
> 
>     Morgan> test.py seems setup for windows but it helps me see what
>     Morgan> you are checking for. I can look at MayaVi's test script
>     Morgan> and see if I can make it work for Linux. In the meantime
>     Morgan> can I run a list by you. Here's an updated Fedora Core 4
>     Morgan> system with only the default repos (+vtk).
> 
>     Morgan> pygtk2-2.6.2-0.fc4.1 vtk-4.5.0-0.cvs.13.fc4
>     Morgan> vtk-python-4.5.0-0.cvs.13.fc4 I need a scipy
>     Morgan> package/install I need a PyOpenGL package/install I need a
>     Morgan> pygtkglext package/install I need a egenix-mx-base
>     Morgan> package/install mysql-4.1.12-2.FC4.1 I need a MySQL-python
>     Morgan> package/install python-matplotlib-0.83.2-2.fc4
> 
> This looks good.  Perhaps you can modify the test script so that it
> works for linux too and post it to the list?
> 
>     Morgan> Is there anything else to test for? Will the newer version
>     Morgan> of mysql be a problem? Is the glade test special or is
>     Morgan> that a check of pygtk as well?  Does pygtk2 provide the
>     Morgan> pygtkglext or is that separate?
> 
> I am not sure how your pygtk packager handles glade -- it may be
> separate.  In debian, we have python2.4-glade2.  pygtkglext is going
> to be separate.  In ubuntu, gtkglext is packaged but pygtkglext is
> not, and so I compiled it separately.  The mysql version should not be
> a problem, I'm running mysql-common 4.0.3.
> 
> FYI, I am running a debian ubuntu package repository at
> 
>   deb http://peds-pc311.bsd.uchicago.edu binary/
> 
> which has vtk patented, scipy cvs, matplotlib cvs, and pbrain with all
> it's dependencies.  In principle you can throw in an ubuntu live CD,
> add the line above to /etc/apt/sources.list and do 
> 
>  > sudo apt-get update
>  > sudo apt-get install pbrain
> 
> I'm writing this mainly for the benefit of the list, but for you as
> well in case you every decide to switch over.  Word of warning: I am a
> total newbie at making debs and didn't really take the time to "do it
> right".  Caveat emptor.
> 
>     Morgan> Hmm. I know import vtk works but VTK patented will be more
>     Morgan> challenging I think. Is there an easy way to check if it
>     Morgan> was built with patented? I have built it from source but
>     Morgan> it is a pain. I have been using the Creatis rpms which
>     Morgan> might have patented included. I don't know. The
>     Morgan> monkeyrpms.net developer might be helping me package
>     Morgan> things the way we need for pbrain and brainstat so this is
>     Morgan> all good info to have no matter what.
> 
> The following should be a good test for patented 
> 
>     >>> import vtk
>     >>> d = vtk.vtkDecimate()
>     >>> c = vtk.vtkMarchingCubes()
> 
> You might want to add it to the test script.  I don't have ready
> access to a system right now that doesn't have vtk patented, so I
> don't know what exceptions will be raised if they are not available.
> 
> We've also been working on some new features for eegview: a
> sophisticated annotation browser, some new statistical features like
> FFT surrogates and phase synchronization measures, so hopefully we can
> get these wrapped up and tested and roll out a new release in the next
> month or so.
> 
> 
> Good luck!  
> JDH
-- 
Morgan Hough  DPhil student
FMRIB, John Radcliffe Hospital, Headington, Oxford OX3 9DU, UK
mh...@fm...  www.fmrib.ox.ac.uk/~mhough
tel +44 (0) 1865 222545  fax +44 (0) 1865 222717

>>>>> "Morgan" == Morgan Hough <mh...@fm...> writes:

    Morgan> test.py seems setup for windows but it helps me see what
    Morgan> you are checking for. I can look at MayaVi's test script
    Morgan> and see if I can make it work for Linux. In the meantime
    Morgan> can I run a list by you. Here's an updated Fedora Core 4
    Morgan> system with only the default repos (+vtk).

    Morgan> pygtk2-2.6.2-0.fc4.1 vtk-4.5.0-0.cvs.13.fc4
    Morgan> vtk-python-4.5.0-0.cvs.13.fc4 I need a scipy
    Morgan> package/install I need a PyOpenGL package/install I need a
    Morgan> pygtkglext package/install I need a egenix-mx-base
    Morgan> package/install mysql-4.1.12-2.FC4.1 I need a MySQL-python
    Morgan> package/install python-matplotlib-0.83.2-2.fc4

This looks good.  Perhaps you can modify the test script so that it
works for linux too and post it to the list?

    Morgan> Is there anything else to test for? Will the newer version
    Morgan> of mysql be a problem? Is the glade test special or is
    Morgan> that a check of pygtk as well?  Does pygtk2 provide the
    Morgan> pygtkglext or is that separate?

I am not sure how your pygtk packager handles glade -- it may be
separate.  In debian, we have python2.4-glade2.  pygtkglext is going
to be separate.  In ubuntu, gtkglext is packaged but pygtkglext is
not, and so I compiled it separately.  The mysql version should not be
a problem, I'm running mysql-common 4.0.3.

FYI, I am running a debian ubuntu package repository at

  deb http://peds-pc311.bsd.uchicago.edu binary/

which has vtk patented, scipy cvs, matplotlib cvs, and pbrain with all
it's dependencies.  In principle you can throw in an ubuntu live CD,
add the line above to /etc/apt/sources.list and do 

 > sudo apt-get update
 > sudo apt-get install pbrain

I'm writing this mainly for the benefit of the list, but for you as
well in case you every decide to switch over.  Word of warning: I am a
total newbie at making debs and didn't really take the time to "do it
right".  Caveat emptor.

    Morgan> Hmm. I know import vtk works but VTK patented will be more
    Morgan> challenging I think. Is there an easy way to check if it
    Morgan> was built with patented? I have built it from source but
    Morgan> it is a pain. I have been using the Creatis rpms which
    Morgan> might have patented included. I don't know. The
    Morgan> monkeyrpms.net developer might be helping me package
    Morgan> things the way we need for pbrain and brainstat so this is
    Morgan> all good info to have no matter what.

The following should be a good test for patented 

    >>> import vtk
    >>> d = vtk.vtkDecimate()
    >>> c = vtk.vtkMarchingCubes()

You might want to add it to the test script.  I don't have ready
access to a system right now that doesn't have vtk patented, so I
don't know what exceptions will be raised if they are not available.

We've also been working on some new features for eegview: a
sophisticated annotation browser, some new statistical features like
FFT surrogates and phase synchronization measures, so hopefully we can
get these wrapped up and tested and roll out a new release in the next
month or so.

Good luck!  
JDH

We've just uploaded the first public release of the pbrain software
which includes eegview and loc3d.  

  Loc3D documentation - http://pbrain.sourceforge.net/Loc3D_README.htm
  Downloads - http://sourceforge.net/projects/pbrain
  Prereqs -  http://pbrain.sourceforge.net/pbrain_win32_dependencies.zip
  Homepage - http://pbrain.sourceforge.net/

As time progresses, we'll work on streamlining the install process and
expanding the documentation.

Hope this is enough to get you started.

Please direct questions to the users mailing list

  http://lists.sourceforge.net/mailman/listinfo/pbrain-users

Cheers!
JDH

2004	Jan	Feb	Mar	Apr	May	Jun	Jul	Aug	Sep	Oct	Nov	Dec (1)
2005	Jan	Feb	Mar	Apr	May	Jun	Jul	Aug (2)	Sep	Oct (8)	Nov (1)	Dec
2006	Jan (3)	Feb	Mar (1)	Apr	May (1)	Jun	Jul	Aug	Sep	Oct	Nov	Dec (1)
2007	Jan	Feb	Mar (1)	Apr	May	Jun	Jul	Aug	Sep	Oct	Nov	Dec
2008	Jan	Feb	Mar (1)	Apr	May	Jun	Jul	Aug	Sep	Oct	Nov	Dec

pbrain-users Mailing List for brain analysis software in python

pbrain-users — General discussion and help for users of pbrain software