This is the mail archive of the
gsl-discuss@sources.redhat.com
mailing list for the GSL project.
Re: Documentation problem
- From: Brian Gough <bjg at network-theory dot co dot uk>
- To: Lukas Dobrek <dobrek at itp dot uni-hannover dot de>
- Cc: gsl-discuss at sources dot redhat dot com
- Date: Fri, 3 May 2002 19:33:19 +0100 (BST)
- Subject: Re: Documentation problem
- References: <20020502130229.A29328@itp.uni-hannover.de>
Lukas Dobrek writes:
> I just report problem. The problem is misleading documentation to
> Simulated Annealing. The part of algoritm description which clames
> that the size of a step is toshed with Bolzman distribution. Is not
> true.
> take_step (r, new_x, params.step_size);
> This part of docs is IMHO not true:
> An initial guess is supplied. At each step, a point is chosen at
> a random distance from the current one, where the random distance
> @math{r} is distributed according to a Boltzmann distribution @tex
> $r = \exp^{-E/kT}$. @end tex @ifinfo r = e^(-E/kT). @end ifinfo
Thanks for the bug report.
> The secound thing. Can one explain my why this is implemented in
> souch a way. I mean why thera are all those memcpy. I was thinking
> to change this into similar way how the rest of gsl is done.
It was the one of the first parts of GSL to be written so it's
historical. It needs some more work to update it with the rest of the
library.
regards
Brian Gough