[Ipopt] IPOPT query

Stefan Vigerske stefan at math.hu-berlin.de
Sat Oct 21 09:00:36 EDT 2017


Hi,

try relaxing the optimality tolerances, e.g., set parameter "tol" 
(https://www.coin-or.org/Ipopt/documentation/node42.html#SECTION000112010000000000000) 
to a rather large value.
If that is not sufficient, then also increase "compl_inf_tol".
If Ipopt now finds a solution that is not feasible enough, then decrease 
"constr_viol_tol".

Stefan

On 10/02/2017 06:52 PM, Bharath Swaminathan wrote:
> Hello
> 
> I am an IPOPT for Matlab user. I am not able to open the forum for posting queries, hence I am mailing you.
> 
> I was wondering how to stop at a feasible solution (inf_pr <= delta) and not necessarily an optimal solution. It is crucial for my work. Please guide.
> 
> Thanks
> 
> Bharath
> 
> 


-- 
http://www.gams.com/~stefan


More information about the Ipopt mailing list