<div dir="ltr"><div><div><div><div><div><div><div><div>Hi again,<br><br></div>I realized that in the LP files I sent to you,<br></div>there are a lot of extra artificial variables (at least one by constraint),<br>whereas in the first log I sent to you, there were only two artifical variables for all constraints.<br>
</div><div>Please excuse me for that, I had to produce results using this initialization, with smaller instances, for another purpose,<br>and I forgot to change the code before generating LP files.<br></div><div>If you need them, I will be able to generate the LP files with two artificial variables next week.<br>
</div><div><br></div>Anyway, SCIPprintStatistics tells that, a 65000 * 20000 LP,<br>with at least 20000 artificial variables is solved in 300 sec at root node.<br><br>Then, during column generation, 6920 of 7200 seconds are spent reoptimising LPs.</div>
<div>There are 35 column generation iterations.<br></div><div>All negative reduced cost columns are added at each iteration (full pricing).<br></div><div><br>In a 7200 seconds run, 223 seconds were spent pricing.<br>Almost all the time was spent solving LPs ...<br>
SCIPprintStatistics output :<br><br>Pricers : ExecTime SetupTime Calls Vars<br> problem variables: 0.00 - 0 0<br> Exact pricer : 223.51 0.01 34 605<br>
<br>LP : Time Calls Iterations Iter/call Iter/sec Time-0-It Calls-0-It<br> primal LP : 6920.86 34 661444 19454.24 95.57 0.00 0<br> dual LP : 54.79 1 35700 35700.00 651.58 0.00 0<br>
</div></div><br></div><div>Column generation has not the time to terminate at root node.<br><br></div></div>Is the LP solved again and again from scratch ?<br></div><div></div><div><br></div>Best regards,<br><br>Xavier Schepler<br>
</div><div class="gmail_extra"><br><br><div class="gmail_quote">2014-02-24 14:03 GMT+01:00 Xavier Schepler <span dir="ltr"><<a href="mailto:xavier.schepler@gmail.com" target="_blank">xavier.schepler@gmail.com</a>></span>:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Dear doctor Timo,<br>
<br>
Thank you very much for your answer.<br>
<br>
Le 23/02/2014 14:36, Timo Berthold a écrit :<div class=""><br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
1. You imposed a time limit of half an hour to SCIP.<br>
</blockquote></div>
I get it ... This is what is causing "LP solver hit time limit" message ...<br>
The fact that this message appeared several times in different rows led me to a misunderstanding.<br>
<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
whenyou create them on your own, you should consider using the specialized<div class=""><br>
versions of linear constraints (setppc, logicor, kanpsack...)<br>
</div></blockquote>
Thanks for your advice, I will create specialized versions of linear constraints.<div class=""><br>
<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
3. You are not using SoPlex as LP solver, but Cplex. Also, you seem to do<br>
some postprocessing with Cplex as standalone MIP solver.<br>
</blockquote></div>
In fact, CPLEX is the LP solver for 1 instances (strange 1.txt), and SoPlex for the 2 others.<div class=""><br>
<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
4. Not to solve a 100kx100k LP within 20 minutes is unusual, unless it is<br>
really ill-conditioned. It would be interesting to see how any LP solver<br>
behaves on it standalone.<br>
Is it possible that you write it to a file and upload it somewhere<br>
<br>
</blockquote></div>
2 LP files can be downloaded at this URL:<br>
<br>
<a href="http://www.filedropper.com/lpstar" target="_blank">http://www.filedropper.com/<u></u>lpstar</a><br>
<br>
Note that, calling<br>
<br>
SCIPprintTransProblem<br>
<br>
produced an empty file (maybe because the problem is not transformed ?), so I used SCIPwriteMIP to produce these files.<br>
<br>
On my computer, CPLEX solves each of these two instances in a little more than 10 minutes,<br>
when used independently in interactive mode.<div class=""><br>
<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
5. It is conspicuous that your own output of initialization time pretty<br>
much fits the difference between this 1/2 hour and the time SCIP needs for<br>
the root LP. So might it be that this is called when SCIP is already<br>
running and we just the output in different order because it comes via<br>
different streams?<br>
So or so, you might want to add a SCIP_CALL( SCIPprintStatistics(scip) )<br>
after SCIP's termination to see where time went within SCIP.<br>
</blockquote></div>
>From what this command outputs, initialization time is not included into SCIP solving time.<br>
<br>
Best regards,<br>
<br>
Xavier<br>
</blockquote></div><br></div>