[MINC-users] fit_stages

[Andrew Janke]

Hi Lisa,

You are no doubt talking about volgenmodel?  If so this argument
defines the fitting stages.

So if your fit config was this:

cai-harold$ cat bin/fit.10-genmodel.conf
# ICBM nlin conf

@conf = (
   {'step' => 32, 'blur_fwhm' => 16, 'iterations' => 20},
   {'step' => 16, 'blur_fwhm' => 8,  'iterations' => 20},
   {'step' => 12, 'blur_fwhm' => 6,  'iterations' => 20},
   {'step' => 8,  'blur_fwhm' => 4,  'iterations' => 20},
   {'step' => 6,  'blur_fwhm' => 3,  'iterations' => 20},
   {'step' => 4,  'blur_fwhm' => 2,  'iterations' => 10},
   {'step' => 2,  'blur_fwhm' => 1,  'iterations' => 10},
   );

Then a -fit_Stages argument of lin,lin,lin,0,0,0,0,1,1,1,1,2,2,2,2,3,3,3 means:

 * do a 'lin' linear fit of all subjects and make a model.
 * do two more of the same above. (lin, lin)

 * do a fit of all subjects using up to line 0 from the fit_config,
repeat 3 more times
     ({'step' => 32, 'blur_fwhm' => 16, 'iterations' => 20},

 * do a fit of all subjects up to line 1 of the fit config x 4
   {'step' => 32, 'blur_fwhm' => 16, 'iterations' => 20},
   {'step' => 16, 'blur_fwhm' => 8,  'iterations' => 20},

....


a

On 20 July 2011 03:59, Lisa F. Akiyama <lrisa87 at uw.edu> wrote:
> Hello MINC experts,
>
> Could someone explain to me what the numerical values signify when
> you determine fit_stages (e.g. lin,lin,lin,0,0,0,0,1,1,1,1,2,2,2,2,3,3,3).
>
> Thank you.
>
>
> Best,
> Lisa
> _______________________________________________
> MINC-users at bic.mni.mcgill.ca
> http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users
>