On Sun, Oct 30, 2011 at 1:53 PM, Ted Ralphs <span dir="ltr"><<a href="mailto:ted@lehigh.edu">ted@lehigh.edu</a>></span> wrote:<br><div class="gmail_quote"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">
<br><br><div class="gmail_quote"><div class="im">On Thu, Oct 13, 2011 at 11:13 AM, Haroldo Santos <span dir="ltr"><<a href="mailto:haroldo.santos@gmail.com" target="_blank">haroldo.santos@gmail.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<font face="arial,helvetica,sans-serif">Hi,<br><br>I've been using SYMPHONY in the last days and I'm really appreciating how it executes well in parallel, great job !<br><br>There are some questions I have about heuristics :<br>
<br>Is there any command line parameter to activate (increase number of passes) of the feasibility pump heuristic ?<br></font></blockquote><div><br></div></div><div>There isn't a way to set the parameters related to the feasibility pump on the command line, but these parameters can easily be set in a parameter file with the "-f" command-line switch. The parameters associated with the feasibility pump are the ones that begine with "fp_". You can find them in master_func.c by searching for the string "fp_" (line 2957 in the trunk version). Does this help?</div>
</div></blockquote><div>Ok, I'll try it. <br></div><blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;"><div class="gmail_quote"><div class="im">
<div> </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><font face="arial,helvetica,sans-serif">Is there any way to inform a file with one (or more) initial feasible solutions (this would be really nice) ?<br>
</font></blockquote><div><br></div></div><div>For what purpose? There is a way to give an initial solution to SYMPHONY, but it's really just used to get an initial bound. If no better solution is found by the end, it will then be reported as the optimum. Otherwise, I don't think it's used for anything.</div>
</div></blockquote><div>There are some MIP heuristics which consider information from a pool of solutions (e.g.: crossing, implemented in CBC). <br><br></div><blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;">
<div class="gmail_quote"><div class="im">
<div><br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><font face="arial,helvetica,sans-serif">Are there plans to implement other heuristics in SYMPHONY (e.g. RINS) ?<br>
</font></blockquote><div><br></div></div><div>Well, there are always plans, but no one to implement them at the moment :). We worked on RINS a few years ago, but it never made it into the code. Actually, it may not be that hard to use Cbc's heuristics within SYMPHONY, although RINS is probably a difficult one to export. If you are interested in doing some work in this area, I can try to shove you in the right direction :).</div>
</div></blockquote><div> Yes, I'm interest in coding some MIP heuristics and integrating them in SYMPHONY would be great. I'll read more about symphony organization and contact you as soon as I have a better understanding of how it works.<br>
<br>Meanwhile, if you have some pointer to indicate me on where to start...<br><br></div><blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;"><div class="gmail_quote">
<div><br></div><div>Cheers,</div><div><br></div><div>Ted</div><div>-- </div></div><font color="#888888">Dr. Ted Ralphs<br>Associate Professor, Lehigh University<br><a href="tel:%28610%29%20628-1280" value="+16106281280" target="_blank">(610) 628-1280</a><br>
ted 'at' lehigh 'dot' edu<br><a href="http://coral.ie.lehigh.edu/%7Eted" target="_blank">coral.ie.lehigh.edu/~ted</a><br>
<br>
</font></blockquote></div><br><br clear="all"><br>-- <br>=============================================================<br>Haroldo Gambini Santos<br>Computing Department - Universidade Federal de Ouro Preto - UFOP<br>email: haroldo [at ] <a href="http://iceb.ufop.br" target="_blank">iceb.ufop.br</a><br>
home/research page: <a href="http://www.decom.ufop.br/haroldo/" target="_blank">www.decom.ufop.br/haroldo/</a><br> <br>"Computer science is no more about computers than astronomy <br>is about telescopes." Edsger Dijkstra<br>
<br>