[Clp] QP methods comparison

armen nahapetyan armen_nahapetyan at yahoo.com
Mon Sep 28 14:25:46 EDT 2009


 Hello everyone,

I tried to use Barrier method for solving QP problems and got a strange message at the end of some QP problems. Clp solves the problem but at the end it reports:
"Unknown objective -234.95 - 26 iterations time 8.542"
But with Clp default QP method, i.e. 'Reduced Gradient' Clp reports normal solution.

Does it mean that Barrier method is not enough robust?

Whether we can rely only on 'Reduces Gradient' for QP problems in general?

Thanks in advance,
Armen




      


      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://list.coin-or.org/pipermail/clp/attachments/20090928/a574fee8/attachment.html>


More information about the Clp mailing list