This is the mail archive of the
gsl-discuss@sources.redhat.com
mailing list for the GSL project.
Re: Error in multifit_lmsder (?)
I've compiled the program but it will take me a while to study it.
Nicolai Hanssing writes:
> Dear listmembers
>
> I have struggeled with the "gsl_multifit_fdfsolver_lmsder", and it works
> fine *once*. I have tried to look over the source of "gsl_multifit_", but
> haven't found anything.
>
> I still think that it's probably my own routine,
> but it follows the example ing gsl-ref closely...
>
> The "lmsder" converges correctly the firste time, but when the routine is
> called again [with the same parameters] it fails, and sends X= ['nan'
> 'nan' 'nan'] to the solverfunctions f and df.
>
> I do allocate space for the solver, initialize the solver, and I do free
> the solver afterwards. And I cant find any reason for the behaivour of the
> solver.
>
> 3 calls as the following fails... [And I dont get it :-)]
> solve_nonlin_sys_multifit_nlin(&v_topp.vector,v_alpha); /* OK */
> solve_nonlin_sys_multifit_nlin(&v_topp.vector,v_alpha); /* FAILS */
> solve_nonlin_sys_multifit_nlin(&v_topp.vector,v_alpha); /* FAILS */
>
> ANy insight will be much appreciated.
>
> The problem solved is n>=3 equations: 0 = P0 - X0*sin(X1*P1+X2)
>
> Regards
> Nicolai Hanssing
> Denmark
>