[Insight-developers] differences in itkMultiResolutionPDEDeformableRegistrationTest results due to portable-rounding and pixel-centeredness

Michel Audette michel.audette at kitware.com
Tue May 26 14:35:27 EDT 2009


Hi Bill,

the recent baseline files predate Tom's latest commit. I have not committed
anything this week, including regression data. They do not seem to be
inconsistent with Tom's fix, however. The number of failing tests with flags
on is the as same as prior to the fix.

Best wishes,

Michel

On Tue, May 26, 2009 at 1:51 PM, Bill Lorensen <bill.lorensen at gmail.com>wrote:

> Did the new baseline include the fix that Tom submitted today, .i.e to
> make the default rounding to be round up?
>
> On Tue, May 26, 2009 at 1:31 PM, Michel Audette
> <michel.audette at kitware.com> wrote:
> > Dear All,
> >
> > among the changes made to account for differences in
> > itkMultiResolutionPDEDeformableRegistrationTest results due to portable
> > rounding and pixel-centeredness, many tests have required a new set of
> > regression baseline data, based usually on boundary effects of a pixel or
> > so. The following has a more significant difference, over a band several
> > pixels wide,
> >
> > itkMultiResolutionPDEDeformableRegistrationTest
> > http://www.cdash.org/CDash/testDetails.php?test=23879063&build=341180
> >
> > due to the influence of the new behavior over many resolution levels. The
> > question with this test is whether to accept the new result as is, and
> > commit it accordingly in baseline data as
> > itkMultiResolutionPDEDeformableRegistrationTestPixelCentered.png, or to
> view
> > this error with suspicion and step through with the debugger?
> >
> > Best wishes,
> >
> > Michel
> >
> > --
> > Michel Audette, Ph.D.
> > R & D Engineer,
> > Kitware Inc.,
> > Chapel Hill, N.C.
> >
> >
> > _______________________________________________
> > 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 ITK FAQ at:
> > http://www.itk.org/Wiki/ITK_FAQ
> >
> > Follow this link to subscribe/unsubscribe:
> > http://www.itk.org/mailman/listinfo/insight-developers
> >
> >
>



-- 
Michel Audette, Ph.D.
R & D Engineer,
Kitware Inc.,
Chapel Hill, N.C.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.itk.org/mailman/private/insight-developers/attachments/20090526/75b66bb4/attachment.htm>


More information about the Insight-developers mailing list