[Insight-developers] Dashboard Looking Good

[Luis Ibanez]

Hi Bill,


The DeformableRegistration15 tests are failing due to a timeout,
not really because they are crashing or producing incorrect
results.


     Each one of these tests need about 1 hour to run.  :-/


Here is (for example) the summary of one of these tests:

   Process                            time(seconds)   memory (Kb)

   Rigid Initialization                  1.10567             0
   Rigid Registration                    4.33729           524
   Affine Registration                  11.2681           2024
   Deformable Registration Coarse       21.4911           4316
   Deformable Registration Fine       3676.61           681116

[These timings are in a Quad-Core processor]



It was my mistake to leave a "normal" timeout of 1,500 seconds
in "redwall"s ctest script.

I'm increasing the timeout to 15,000 seconds and rerunning the
Nightly. This should do the trick... but we will only know for
sure by 10:00am ~ 10:30 today.


These tests are quite demanding, but they are also very important
because they exercise the 3D deformable registration with a full
pipeline that includes:

   a) use of Oriented images
   b) centered transform initialization
   c) rigid registration
   d) affine registration
   e) bspline coarse registration
   f) bspline fine registration
   g) resampling
   h) generation of deformation field


Somehow these test are a response of the many things that
users forget to do right when putting together a registration
program.

I'll condition the tests to run only on machines that have
setup a TIMEOUT larger than 10,000 seconds.


      Luis


---------------------
Bill Lorensen wrote:
> Luis,
> 
> The changes you made to the multi res registration examples are
> producing great results. Looks like we need a second regression image
> for the second multi res registration example.
> 
> The tests DeformableRegistration15Test1 and
> DeformableRegistration15Test2 are failing on all platforms they run
> on. Looks like they use the data from BrainWeb (Test1) or Kilian
> (Test2).
> 
> It's not clear to me what BrainWeb data is required to run the tests.
> And I don't think Kilian's data should be in a test checked into ITK
> unless we can add his data to the ITK distribution.
> 
> Bill
>