[Insight-users] Optimizing on a subset of the metric's parameters

[Aviv Hurvitz]

Hi all.


I have a metric class which I defined.


 I would like use itk::AmoebaOptimizer to search through only a subset of
the metric's parameters space. For example, I want the search to only
perturb the first two parameters, and to keep the other parameters
constantly on their initial values.


 I was wondering what is the best way to program this. I tried to set the
"simplex delta" field of the Amoeba optimizer to 0's on the parameters that
I want to keep constant. Unfortunately, the optimizer still attempts to
perturb them.


 Another solution, which must work, is to wrap my metric in another metric
whose sole purpose is to remap the smaller parameters vector into the full
parameters vector. I was going to do this, but then I thought I'm probably
reinventing something in an unwieldy way. Is there a better solution for
this problem?


 Kind regards,


 - Aviv


 ----

Aviv Hurvitz
Graduate Student
School of Computer Science and Engineering
The Hebrew University of Jerusalem
Jerusalem 91904, Israel
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://public.kitware.com/pipermail/insight-users/attachments/20070118/ed28f799/attachment.htm