[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Help-gsl] non-linear least squares fitting
From: |
Joakim Hove |
Subject: |
Re: [Help-gsl] non-linear least squares fitting |
Date: |
Fri, 08 Apr 2005 12:44:51 +0200 |
User-agent: |
Gnus/5.1006 (Gnus v5.10.6) Emacs/21.2 (gnu/linux) |
<Disclaimer>
These ideas are all untested
</Disclaimer>
I would recomend creating struct containing the relevant flags and
default values, and then pass a pointer to such an instance as the
void parameter in the fitting routines - that way you avoid the global
variables.
To ensure[1] that the parameters are not changed I would set the
jacobian for those parameters which should be fixed forcibly to zero.
[1]: I don't know the details of the algorithm, this might not be
sufficient; and it is probably *not* optimal.
> PS. The manual text on the non-linear least squares fitting follows from the
> section on multidimensional minimization. But in this case, I find the
> notion of parameters and x values to be backwards. Most would expect "x" to
> be the name of the data point, and not the parameter to be minimized.
I fully agree :-)
HTH - Joakim
--
Joakim Hove
hove AT ift uib no /
Tlf: +47 (55 5)8 27 90 / Stabburveien 18
Fax: +47 (55 5)8 94 40 / N-5231 Paradis
http://www.ift.uib.no/~hove/ / 55 91 28 18 / 92 68 57 04