[Ipopt] Start point for derivatives checking

[Stefan Vigerske]

Hi,

according to http://www.coin-or.org/Ipopt/documentation/node53.html :
"The verification is done by a simple finite differences approximation, 
where each component of the user-provided starting point is perturbed 
one of the other. The relative size of the perturbation is determined by 
the option derivative_test_perturbation. The default value ($ 10^{-8}$, 
about the square root of the machine precision) is probably fine in most 
cases,..."

Are you sure its taken completely random?

Stefan

On 02/19/2013 06:34 PM, Xin Zhao wrote:
> Dear all,
>
>
> Sorry to bother you all again.
>
> Is there any way to specify the start point for derivatives checking? I've
> noticed that the start point for the derivatives checking seems to be taken
> randomly. Is there a way to force start point to be the one implemented in
> get_start_point()?
>
> Thanks.
>
>
> Best,
> Xin
>
>
>
> _______________________________________________
> Ipopt mailing list
> Ipopt at list.coin-or.org
> http://list.coin-or.org/mailman/listinfo/ipopt
>