[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: lsqcurvefit alternative
From: |
Olaf Till |
Subject: |
Re: lsqcurvefit alternative |
Date: |
Wed, 28 Feb 2007 13:38:01 +0100 |
User-agent: |
Mutt/1.5.9i |
On Wed, Feb 28, 2007 at 04:01:50PM +1300, Ana Rakonjac wrote:
> On 2/27/07, Henry F. Mollet <address@hidden> wrote:
> > Will below help?
> > Henry
> > octave:1> help leasqr
> >
> > leasqr is the user-defined function from the file
> > /usr/local/share/octave/2.1.71/site/m/octave-forge/optim/leasqr.m
> >
> > function [f,p,kvg,iter,corp,covp,covr,stdresid,Z,r2]=
> > leasqr(x,y,pin,F,{stol,niter,wt,dp,dFdp,options})
> >
> > Levenberg-Marquardt nonlinear regression of f(x,p) to y(x).
> > Cut details
> >
> >
> > on 2/26/07 2:44 PM, Ana Rakonjac at address@hidden wrote:
> >
>
>
> Unfortunately that doesn't help. That's some other optimisation
> routine that Matlab has as well.
Maybe you are mistaking that for 'lsqr'.
Matlabs lsqcurvefit lets you choose between 3 optimization methods,
one of which is the Levenberg-Marquardt method. The latter is provided
by octaves 'leasqr'.
Olaf