[Coin-discuss] Cbc questions

[John J Forrest]

Brady,

It was too restricting to make the infeasibility be between 0.0 and 0.5 as 
it would mean that all types of discontinuity objects would have to know 
about the others to get the scaling right AND make sure that one with 
largest infeasibility was taken e.g. what about SOS?  I could get round to 
changing names and doing a new "infeasibility" or I could add a method 
with a new name which would be correct for integer variables.

As to the other point, I could dig deeper but I got into trouble when I 
tried to give statistics (switched on by branchAndBound(n) where n is 
0,1,2 or 3) as node numbers refused to line up correctly.

If you are looking at the tree then the code to print out statistics does 
store values of integer variables.  So look at the output you get from 
branchAndBound(3) and I "may" correct bugs if it is simple to do so. 
Anyway tell me what is wrong with the output.

John



Brady Hunsaker <hunsaker at engr.pitt.edu> 
Sent by: coin-discuss-bounces at list.coin-or.org
02/17/2006 12:07 PM
Please respond to
Discussions about open source software for Operations Research 
<coin-discuss at list.coin-or.org>


To
coin-discuss at list.coin-or.org
cc

Subject
[Coin-discuss] Cbc questions






I'm doing some work that includes examining the tree created by CBC.

John and Robin, I expect that the two of you are most likely able to
answer my questions.

First, one of the data points that I am looking at is the sum of the
infeasibilities at a subproblem.  My concern is that the value returned
by CbcSimpleIntegerDynamicPseudoCost::infeasibility (in
CbcBranchDynamic.cpp) does not return a value between 0 and 0.5, as the
comments seem to suggest.

Is this intentional?


Second, in CbcModel.cpp, around line 1060 there is a comment that says
"set node number correctly".  The accompanying call to setNodeNumber
actual changes the node number, however, and the increment of
nodeNumber2_ a few lines later also seems to be wrong.

By commenting both of these out I get a consistent node numbering.  Is
that correct?  Admittedly, CBC keeps track of nodes in a different way
then I usually think about, since it doesn't create a node until it is
processed (and only if it will have children).  It may be that someone
added those two lines while struggling as I did with this issue.

Could you clarify?

Thanks,
Brady




-- 
Brady Hunsaker
Assistant Professor
Industrial Engineering
University of Pittsburgh
http://www.engr.pitt.edu/hunsaker/
_______________________________________________
Coin-discuss mailing list
Coin-discuss at list.coin-or.org
http://list.coin-or.org/mailman/listinfo/coin-discuss

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://list.coin-or.org/pipermail/coin-discuss/attachments/20060217/efcc9a20/attachment.html>