[Ipopt] Ipopt tuning for Quadratic Problem

Andreas Waechter andreasw at watson.ibm.com
Fri Feb 26 16:52:09 EST 2010


Hallo Christian,

Do you have an AMPL model that you can email me to reproduce the 
exception?

Did you try to increase bound_push and bound_frac?

Sorry for the delay in my response.  I don't have much time for Ipopt 
maintenance these days....

Andreas

On Tue, 9 Feb 2010, Schulz, Christian wrote:

> Hi,
> we are trying to use Ipopt 3.8.0 for a convex quadratic problem.
> We observed the following:
> 1) mehrotra_algorithm yes, does not work, i.e. causes an exception after some steps
> 2) changing mu_startegy to adaptive does not result in a significantly speed up
> Are there any default values to get the most out of Ipopt for convex quadratic problem?
> Thank You,
> Christian
>



More information about the Ipopt mailing list