[MINC-users] failures when running N3 with small smoothingdistances

[Vitali Zagorodnov]

I am using N3 version 1.1 (the one that comes with FreeSurfer). Not sure what ebtks refers to, I am using N3 as part of FreeSurfer package, so not sure what version they are using.

Maybe somebody who has the very latest versions installed can replicate these errors?

Vitali

-----Original Message-----
From: minc-users-bounces at bic.mni.mcgill.ca [mailto:minc-users-bounces at bic.mni.mcgill.ca] On Behalf Of Andrew Janke
Sent: Monday, May 04, 2009 10:33 PM
To: MINC users mailing list
Subject: Re: [MINC-users] failures when running N3 with small smoothingdistances

Hi Vitali,

2009/5/5 Vitali Zagorodnov <ZVitali at ntu.edu.sg>:
> Hi, I am having two specific failures (see below) when running N3 with
> small smoothing distances, <30mm. Has anybody else experience similar
> errors? Are these limitations of the algorithm or a bug?
>
> 1)      Smoothing distance =20mm on some brains only. The error occurs
> after iterations are over.
>
> Not implemented yet in cache_volume_range_has_changed()

These warnings are harmless.

> inputCompactField(): Incorrect number of coefficients

This not so harmless. :)  In this case it was an esoteric bug that was
squashed in ebtks a while back.

> 2)      Smoothing distance =10mm on all brains tested. The error occurs
> even before first iteration:
>
> nu_estimate_np_and_em: crashed while running spline_smooth (termination
> status=6)

This can mean a few things but I would first suggest that you use the
latest versions of ebtks and N3. Which versions of these are you
currently running and an which platform?


--
Andrew Janke
(a.janke at gmail.com || http://a.janke.googlepages.com/)
Canberra->Australia    +61 (402) 700 883

_______________________________________________
MINC-users at bic.mni.mcgill.ca
http://www2.bic.mni.mcgill.ca/mailman/listinfo/minc-users