Re: [BLAST_ANAWARE] convergence of Newton fitter, and chi^2

From: Frederick T. Lee (tong@mit.edu)
Date: Tue Jun 03 2003 - 22:59:36 EDT


I believe that lrn.C will be disappearing gradually and
we will have DST class for analyzing data.
I think Chi is working on it now. Right?
I agree with Chi on that root script should be short
and easy to understand. Long script makes it very
hard to follow and easy to make mistake.

-T

vitaliy ziskin wrote:
>
> Chris,
> This is all great, but we all use lrn.C to analyzed data. And by
> definition if lrn.C does not work with newer verstions of BlastLib (i.e
> greater than 2.18) it is useless to most of us. And my experience is
> that lrn.C does not work with new versions. Can you please make sure
> that lrn.C works with the version that you are working on. Of course
> the best would be if most of what lrn.C has to offer is part of the
> standart v3 release.
>
>
> Cheers, Vitaliy
>
> Chris Crawford wrote:
>
> > hi tong,
> > i'm almost done v3. there is something checked in there, but it
> > still uses v2 geometry principally. you might have to use the -A flag.
> > i'm also tagging v2_20 soon, which will have wang's fast fitter.
> > --chris
> >
> > Frederick Tong-Uk Lee wrote:
> >
> >> Chris,
> >>
> >> This sounds good.
> >> Hope we get the resolution what we need.
> >>
> >> Hey, by the way, are you doing something with v3?
> >> We are trying to cvs out Aaron's v3 MC and
> >> getting empty directories. Do I need to use v2_19
> >> for v3 or something?
> >>
> >> Thanks,
> >> -T
> >>
> >

-- 
Frederick Tong-Uk Lee               	Senior Research Associate
Laboratory of Nuclear Science at MIT	tong@lns.mit.edu
26-402                              	tel:  (617) 258-5437
77 Massachusetts                    	fax:  (617) 258-5440
Cambridge, MA 02139



This archive was generated by hypermail 2.1.2 : Mon Feb 24 2014 - 14:07:29 EST