<html>
<head>
<style>
.hmmessage P
{
margin:0px;
padding:0px
}
body.hmmessage
{
font-size: 10pt;
font-family:Verdana
}
</style>
</head>
<body class='hmmessage'>
Thanks Andreas. Thanks Stefan. I got it! :)<BR>
<BR>
Now I have another problem. Most of my problems ends with a problem of restoration, and usually there is a high dual infeasibility.<BR>
<BR>
The derivative checker does not find errors, and I tried to solve the problem directly from an .nl file with the same "results". <BR>
I suspect that I have to tune some parameters to solve my model. The model is not very hard, it is quadratic but not convex, so I'm just trying to find a local optima. Also some times I have good (primal) starting point (read good as point close to an optimum), which I would expect few iterations to converge but it fails. Any recomendation of which parameters are the most helpful?<BR>
<BR>
Thanks again!<BR>
Adela<BR>
<BR>
PS: Example of the output I get:<BR>
<BR>
WARNING: Problem in step computation; switching to emergency mode.<BR>Restoration phase is called at point that is almost feasible,<BR> with constraint violation 1.674242e-10. Abort.<BR><BR>
<BR>Number of Iterations....: 87<BR>
(scaled) (unscaled)<BR>Objective...............: 1.7361700482095586e+01 1.7361700482095586e+01<BR>Dual infeasibility......: 1.5927196841453206e+14 1.5927196841453206e+14<BR>Constraint violation....: 2.2204460492503131e-16 2.2204460492503131e-16<BR>Complementarity.........: 8.6335610358611384e-03 8.6335610358611384e-03<BR>Overall NLP error.......: 8.6335610358611384e-03 1.5927196841453206e+14<BR><BR>
<BR>
<BR>
<BR>
<BR>
<BR>
<BR>
<BR>
<BR>
<BR> <BR>> Date: Wed, 27 May 2009 11:20:32 -0400<BR>> From: andreasw@watson.ibm.com<BR>> To: stefan@math.hu-berlin.de<BR>> CC: pages25300@hotmail.com; ipopt@list.coin-or.org<BR>> Subject: Re: [Ipopt] Iteration info<BR>> <BR>> Adela,<BR>> <BR>> I thought you wanted information within each iteration of Ipopt.<BR>> <BR>> You can get information about the final point from the SolveStatistics <BR>> object. After the optimization, it has additional statistics (such as <BR>> final value of objective function, final scaled and unscaled constraint <BR>> violation etc), see the IpSolveStatistics.hpp file. After the <BR>> optimization, you can get access to this object from the IpoptApplication, <BR>> using the Statistics() method.<BR>> <BR>> Hope this helps,<BR>> <BR>> Andreas<BR>> <BR>> On Wed, 27 May 2009, Stefan Vigerske wrote:<BR>> <BR>> > Dl wrote:<BR>> >><BR>> >><BR>> >> Yes, that's great!<BR>> >><BR>> >><BR>> >><BR>> >> But how can I call for example "curr_constraint_violation()"? Calling ip_cq.curr_constraint_violation() from the ::finalize_solution does not work ... what am I missing?<BR>> ><BR>> > ip_cq->curr_constraint_violation()<BR>> ><BR>> > and you need to include the IpIpoptCalculatedQuantities.hpp.<BR>> ><BR>> > But this file is not installed by Ipopt by default.<BR>> > You need the following additional hpp-files from the Ipopt source when<BR>> > using IpoptCalculatedQuantities:<BR>> > $(IPOPTSRCDIR)/src/Algorithm/IpIpoptCalculatedQuantities.hpp \<BR>> > $(IPOPTSRCDIR)/src/Algorithm/IpIpoptNLP.hpp \<BR>> > $(IPOPTSRCDIR)/src/Algorithm/IpIpoptData.hpp \<BR>> > $(IPOPTSRCDIR)/src/Algorithm/IpNLPScaling.hpp \<BR>> > $(IPOPTSRCDIR)/src/Algorithm/IpIteratesVector.hpp \<BR>> > $(IPOPTSRCDIR)/src/Algorithm/IpTimingStatistics.hpp \<BR>> > $(IPOPTSRCDIR)/src/LinAlg/IpScaledMatrix.hpp \<BR>> > $(IPOPTSRCDIR)/src/LinAlg/IpSymScaledMatrix.hpp \<BR>> > $(IPOPTSRCDIR)/src/LinAlg/IpCompoundVector.hpp<BR>> ><BR>> > Stefan<BR>> ><BR>> > -- <BR>> > Stefan Vigerske<BR>> > Humboldt University Berlin, Numerical Mathematics<BR>> > http://www.math.hu-berlin.de/~stefan<BR>> ><BR>> > _______________________________________________<BR>> > Ipopt mailing list<BR>> > Ipopt@list.coin-or.org<BR>> > http://list.coin-or.org/mailman/listinfo/ipopt<BR>> ><BR><br /><hr />Windows Live™: Keep your life in sync. <a href='http://windowslive.com/explore?ocid=TXT_TAGLM_WL_t1_allup_explore_012009' target='_new'>Check it out!</a></body>
</html>