headers
Bastil2001@yahoo.de | 1 Mar 2009 10:08
Picon
Favicon

Feature Extraction Plans

Dear group,

I have seen some notes about plans to include feature extraction into
Paraview (http://www.paraview.org/paraview/project/News.html#feature).
Can anybody give me some details about plans and schedules? I am very
interested in that. Thanks.

BastiL
_______________________________________________
Powered by www.kitware.com

Visit other Kitware open-source projects at http://www.kitware.com/opensource/opensource.html

Please keep messages on-topic and check the ParaView Wiki at: http://paraview.org/Wiki/ParaView

Follow this link to subscribe/unsubscribe:
http://www.paraview.org/mailman/listinfo/paraview
Permalink | Reply | 2 comments |
headers
Ricardo Reis | 2 Mar 2009 01:15
Picon
Favicon
Gravatar

Re: Handshake failed :: but I'm using exactly the same CVS

On Sat, 28 Feb 2009, Utkarsh Ayachit wrote:

> Add the line before the returns as you have shown.

- I have added the lines in both server and client machine.
- make clean and then compiled with DEBUG
- error is gone
- make clean and compile with RELEASE (-O3 -march=native) error is gone
I did the former tests from the compile dir

then I've done make install to the install dir and the error is gone

- Then I did a fresh cvs update and compiled it in the server machine and 
in a client machine and it's all ok

So, I'm baffled and don't understand why but everyting seems to be working 
ok.

  thanks for your trouble

  Ricardo Reis

  'Non Serviam'

  PhD student  <at>  Lasef
  Computational Fluid Dynamics, High Performance Computing, Turbulence
  http://www.lasef.ist.utl.pt

  &
(Continue reading)

Permalink | Reply | 0 comments |
headers
Luc Bordier | 2 Mar 2009 11:06

CellDataToPointData in paraview script

Hello,

I'm starting to use paraview in script mode using pvpython.

I succeeded to load my data and to make a representation of my U 
CellData fields using a Cut() filters.
This works fine.

But now I would like to have the magnitude of my U field for PointData 
and I'm stuck in the use of CellDataToPointData filter.
The following script only give me a blank field for my Cut() when 
passing trough the CellDataToPointData filter.

Surely I forgot something (interpolation, choice of component of field), 
but even by looking at the dir() and help() I didn't manage to have my 
field interpolated at points and represented in colors.

Thanks for anyone who could give me hints to solve this problem.

python script :

con = servermanager.Connect()
vtkf = 
servermanager.sources.LegacyVTKFileReader(FileNames="VTK\\design71_1.vtk")
vtkf.UpdatePipeline()

implicit_functions = servermanager.createModule("implicit_functions")
plane = implicit_functions.Plane()
plane.Origin = [0,0,-0.5]
plane.Normal = [0,0,1]
(Continue reading)

Permalink | Reply | 5 comments |
headers
Pawel Krupinski | 2 Mar 2009 11:35
Picon
Favicon

Glyph scaling and orienting

Hi,
I have got a data which I would like to display as ellipsoids of different sizes and orientations. I was thinking about getting ellipsoids as sphere glyphs scaled by components of the vector data, but than I need to orient them according to another vector. However you can specify only one vector for scaling and/or orienting the glyphs. Does anybody know a way (or workaround) for doing this?
Thanks.


_______________________________________________
Powered by www.kitware.com

Visit other Kitware open-source projects at http://www.kitware.com/opensource/opensource.html

Please keep messages on-topic and check the ParaView Wiki at: http://paraview.org/Wiki/ParaView

Follow this link to subscribe/unsubscribe:
http://www.paraview.org/mailman/listinfo/paraview
Permalink | Reply | 1 comment |
headers
mirko heuegger | 2 Mar 2009 13:13
Picon

Re: Glyph scaling and orienting

Hello!

On Mon, Mar 2, 2009 at 11:35 AM, Pawel Krupinski <exmeson <at> yahoo.com> wrote:

> - [ cut ] -
> - [  glyphs scaled by components of vector data and glyph-orientation
>      by another vector.  ] -
> Does anybody know a way (or workaround) for doing this?

I suggest that you 'convert' one vector data into a scalar (e.g. over
the Calculator-filter, and by using the functions: mag(<vector>),
iHat.<vector>, jHat.<vector>, or kHat.<vector> (magnitude, X, Y, and Z
of a vector) for calculating the desired vector-parts).
Then change the 'Scale Mode', of the Glyph-filter, to 'scalar', and
select the calculation-result as scalar-field.

hth

mirko

--

-- 
Real programmers don't document; if it was
hard to write, it should be hard to understand.
_______________________________________________
Powered by www.kitware.com

Visit other Kitware open-source projects at http://www.kitware.com/opensource/opensource.html

Please keep messages on-topic and check the ParaView Wiki at: http://paraview.org/Wiki/ParaView

Follow this link to subscribe/unsubscribe:
http://www.paraview.org/mailman/listinfo/paraview
Permalink | Reply | 0 comments |
headers
Berk Geveci | 2 Mar 2009 14:14
Favicon

Re: mpirun error

The building instructions are here:

http://paraview.org/Wiki/ParaView:Build_And_Install

You can avoid the Qt dependency if you set PARAVIEW_BUILD_QT_GUI when
configuring with CMake.

-berk

On Fri, Feb 27, 2009 at 2:31 PM, Jim Montine <jmontine <at> sgi.com> wrote:
> Hi,
>
> I downloaded  the source fom the paraview website (paraview-3.4.0.tar.gz).
> What are the smallest number
> of steps I would have to take to build pvserver with MPI?   Does it really
> amount to building everything in the
> source tree?  I was hoping to build jus what I needed and then use the
> prebuilt stuff I already downloaded.
>
>
> Thanks
> ________________________________
> From: Berk Geveci [mailto:berk.geveci <at> kitware.com]
> Sent: Wed 2/25/2009 5:26 PM
> To: Jim Montine; ParaView Mailing List
> Subject: Re: [Paraview] mpirun error
>
> Hi Jim,
>
> The binaries we create are not compiled with MPI (because there are
> many different MPI implementations). If you want to run ParaView with
> MPI, you have to compile it yourself.
>
> -berk
>
> On Wed, Feb 25, 2009 at 7:54 PM, Jim Montine <jmontine <at> sgi.com> wrote:
>> Hi,
>>
>> I simply untarred and ran the server application (3.4.0) on my x86_64
>> linux
>> cluster.
>>
>> I did not build it myself.  Is there a particular type of mpi I should be
>> using (hpmpi, mpich, etc.)
>> for the prebuilt version.
>>
>> Jim
>> ________________________________
>> From: Berk Geveci [mailto:berk.geveci <at> kitware.com]
>> Sent: Wed 2/25/2009 1:56 PM
>> To: Jim Montine
>> Cc: paraview <at> paraview.org
>> Subject: Re: [Paraview] mpirun error
>>
>> Probably a dumb question but is the server compiled with MPI support?
>> What do you see if you create a sphere and then apply process id
>> scalars?
>>
>> -berk
>>
>> On Wed, Feb 25, 2009 at 3:29 PM, Jim Montine <jmontine <at> sgi.com> wrote:
>>> Hi,
>>>
>>> I am able to run pvserver on my Linux cluster and paraview (3.4.0) on my
>>> PC.  I am
>>> able to connect to the server and I can calculate and display a contour
>>> (note, the server
>>> has no gfx, so there is no remote rendering).
>>>
>>> Next I tired:
>>>
>>> mpirun -np pvserver
>>>
>>> And again I can connect from paraview on the PC and calculate and display
>>> the contour.
>>>
>>> However, when I now exit paraview, pvserver on the cluster displays the
>>> following message:
>>>
>>> Client connection closed.
>>> MPI: could not run executable (case #3)
>>> MPI: No details available, no log files found
>>> Killed
>>>
>>> It only displays this as it exits.  I guess this means the server is not
>>> doing any
>>> parallel work - is there a way to tell if the server is working correctly
>>> in
>>> parallel mode?
>>>
>>> What might this problem be casued by?
>>>
>>>
>>>
>>> Thanks
>>> _______________________________________________
>>> Powered by www.kitware.com
>>>
>>> Visit other Kitware open-source projects at
>>> http://www.kitware.com/opensource/opensource.html
>>>
>>> Please keep messages on-topic and check the ParaView Wiki at:
>>> http://paraview.org/Wiki/ParaView
>>>
>>> Follow this link to subscribe/unsubscribe:
>>> http://www.paraview.org/mailman/listinfo/paraview
>>>
>>>
>>
>
_______________________________________________
Powered by www.kitware.com

Visit other Kitware open-source projects at http://www.kitware.com/opensource/opensource.html

Please keep messages on-topic and check the ParaView Wiki at: http://paraview.org/Wiki/ParaView

Follow this link to subscribe/unsubscribe:
http://www.paraview.org/mailman/listinfo/paraview
Permalink | Reply | 0 comments |
headers
Berk Geveci | 2 Mar 2009 21:30
Favicon

Re: Feature Extraction Plans

This work is at a prototype stage right now. Assuming that we are
funded for the next phase, I'd expect we would have initial production
results in 2 years or so. We are also exploring other possibilities to
add feature extraction support to ParaView. I am sorry I can't be more
specific; everything is at the planning stage.

-berk

On Sun, Mar 1, 2009 at 4:08 AM, Bastil2001 <at> yahoo.de <Bastil2001 <at> yahoo.de> wrote:
> Dear group,
>
> I have seen some notes about plans to include feature extraction into
> Paraview (http://www.paraview.org/paraview/project/News.html#feature).
> Can anybody give me some details about plans and schedules? I am very
> interested in that. Thanks.
>
> BastiL
> _______________________________________________
> Powered by www.kitware.com
>
> Visit other Kitware open-source projects at http://www.kitware.com/opensource/opensource.html
>
> Please keep messages on-topic and check the ParaView Wiki at: http://paraview.org/Wiki/ParaView
>
> Follow this link to subscribe/unsubscribe:
> http://www.paraview.org/mailman/listinfo/paraview
>
_______________________________________________
Powered by www.kitware.com

Visit other Kitware open-source projects at http://www.kitware.com/opensource/opensource.html

Please keep messages on-topic and check the ParaView Wiki at: http://paraview.org/Wiki/ParaView

Follow this link to subscribe/unsubscribe:
http://www.paraview.org/mailman/listinfo/paraview
Permalink | Reply | 1 comment |
headers
Jean Favre | 3 Mar 2009 00:10
Picon
Favicon

change of syntax


The following python code works fine in pv3.4

rep = servermanager.CreateRepresentation(solid, view)
rep.ColorArrayName = 'Temperature'

but gives me an error with the cvs version

   rep.ColorArrayName = 'Temperature'
 File
"/local/apps/ParaView3Build/Utilities/VTKPythonWrapping/paraview/servermanager.py",
line 203, in __setattr__
   setter(self, value)
 File
"/local/apps/ParaView3Build/Utilities/VTKPythonWrapping/paraview/servermanager.py",
line 2308, in setProperty
   return self.SetPropertyWithName(propName, value)
 File
"/local/apps/ParaView3Build/Utilities/VTKPythonWrapping/paraview/servermanager.py",
line 252, in SetPropertyWithName
   prop.SetData(arg)
 File
"/local/apps/ParaView3Build/Utilities/VTKPythonWrapping/paraview/servermanager.py",
line 663, in SetData
   for a in self.Available:
 File
"/local/apps/ParaView3Build/Utilities/VTKPythonWrapping/paraview/servermanager.py",
line 519, in __getattr__
   return getattr(self.SMProperty, name)
AttributeError: Available

I must have missed a change being put in. Does someone have a hint?
TIA

_______________________________________________
Powered by www.kitware.com

Visit other Kitware open-source projects at http://www.kitware.com/opensource/opensource.html

Please keep messages on-topic and check the ParaView Wiki at: http://paraview.org/Wiki/ParaView

Follow this link to subscribe/unsubscribe:
http://www.paraview.org/mailman/listinfo/paraview
Permalink | Reply | 0 comments |
headers
Pawel Krupinski | 3 Mar 2009 09:41
Picon
Favicon

Re: Glyph scaling and orienting/programmable filter?

Thanks for reply,
but I don't think it will work in my case. Maybe I wasn't specific enough about my data and it was confusing.
I have point data witch represent centers of ellipsoidal cells. With each point there are associated lengths of ellipsoid axes (e.g. as a vector data), which can be different for each ellipsoid, and directions of those axes in space (3 vectors). I want to display those ellipsoids oriented and in appropriate scale.
So I was thinking:  I need to scale spheres anisotropically to get ellipsoids so I really need to do it by scaling with vector components of the scales/axes_length vector. Scaling by scalars works isotropically so it won't help. Now I need to align axes of each ellipsoid in space. I probably need to do it by programmable filter, but I don't know how to operate on separate glyphs (ellipsoids) and their data. Did anyone have similar problem and solved it somehow?
Thanks.
Pawel 

 >mirko heuegger wrote
>I suggest that you 'convert' one vector data into a scalar (e.g. over
>the Calculator-filter, and by using the functions: mag(<vector>),
>iHat.<vector>, jHat.<vector>, or kHat.<vector> (magnitude, X, Y, and Z
>of a vector) for calculating the desired vector-parts).
>Then change the 'Scale Mode', of the Glyph-filter, to 'scalar', and
>select the calculation-result as scalar-field.

_______________________________________________
Powered by www.kitware.com

Visit other Kitware open-source projects at http://www.kitware.com/opensource/opensource.html

Please keep messages on-topic and check the ParaView Wiki at: http://paraview.org/Wiki/ParaView

Follow this link to subscribe/unsubscribe:
http://www.paraview.org/mailman/listinfo/paraview
Permalink | Reply | 627 comments |
headers
David E DeMarle | 3 Mar 2009 17:14
Favicon
Gravatar

Re: Glyph scaling and orienting/programmable filter?

Try VTK's vtkTensorGlyph
http://www.vtk.org/doc/nightly/html/classvtkTensorGlyph.html filter.
You will have to write a plugin to let paraview use it. See
http://www.paraview.org/Wiki/Plugin_HowTo#Enabling_a_filter_in_VTK

On Tue, Mar 3, 2009 at 3:41 AM, Pawel Krupinski <exmeson <at> yahoo.com> wrote:
> Thanks for reply,
> but I don't think it will work in my case. Maybe I wasn't specific enough
> about my data and it was confusing.
> I have point data witch represent centers of ellipsoidal cells. With each
> point there are associated lengths of ellipsoid axes (e.g. as a vector
> data), which can be different for each ellipsoid, and directions of those
> axes in space (3 vectors). I want to display those ellipsoids oriented and
> in appropriate scale.
> So I was thinking:  I need to scale spheres anisotropically to get
> ellipsoids so I really need to do it by scaling with vector components of
> the scales/axes_length vector. Scaling by scalars works isotropically so it
> won't help. Now I need to align axes of each ellipsoid in space. I probably
> need to do it by programmable filter, but I don't know how to operate on
> separate glyphs (ellipsoids) and their data. Did anyone have similar problem
> and solved it somehow?
> Thanks.
> Pawel
>
>  >mirko heuegger wrote
>>I suggest that you 'convert' one vector data into a scalar (e.g. over
>>the Calculator-filter, and by using the functions: mag(<vector>),
>>iHat.<vector>, jHat.<vector>, or kHat.<vector> (magnitude, X, Y, and Z
>>of a vector) for calculating the desired vector-parts).
>>Then change the 'Scale Mode', of the Glyph-filter, to 'scalar', and
>>select the calculation-result as scalar-field.
>
> _______________________________________________
> Powered by www.kitware.com
>
> Visit other Kitware open-source projects at
> http://www.kitware.com/opensource/opensource.html
>
> Please keep messages on-topic and check the ParaView Wiki at:
> http://paraview.org/Wiki/ParaView
>
> Follow this link to subscribe/unsubscribe:
> http://www.paraview.org/mailman/listinfo/paraview
>
>

--

-- 
David E DeMarle
Kitware, Inc.
R&D Engineer
28 Corporate Drive
Clifton Park, NY 12065-8662
Phone: 518-371-3971 x109
_______________________________________________
Powered by www.kitware.com

Visit other Kitware open-source projects at http://www.kitware.com/opensource/opensource.html

Please keep messages on-topic and check the ParaView Wiki at: http://paraview.org/Wiki/ParaView

Follow this link to subscribe/unsubscribe:
http://www.paraview.org/mailman/listinfo/paraview
Permalink | Reply | 0 comments |

Gmane