<html>
  <head>
    <meta content="text/html; charset=utf-8" http-equiv="Content-Type">
  </head>
  <body bgcolor="#FFFFFF" text="#000000">
    <div class="moz-cite-prefix">Marco,<br>
      <br>
      Just switch Gomory i.e. -gomory off -gmi ifmove <br>
      <br>
      Will be just the same as next release.  You may wish to make it
      -gmi root.<br>
      <br>
      John Forrest<br>
      <br>
      <br>
      On 24/03/15 11:18, mg wrote:<br>
    </div>
    <blockquote
cite="mid:CAOayR=b+Z7ZQtxCR91_twpp+qRQNk_VieCcB4Mqd25Zg1fkZTQ@mail.gmail.com"
      type="cite">
      <div dir="ltr">
        <div>John, thanks a lot for the fast reply :)</div>
        <div><br>
        </div>
        <div>
          <div>Do you think I can still use the 2.9.3 version (release)
            and just turn gomory cuts off ?</div>
          <div>Does this version (2.9.3) suffer of other known problems
            that may be worth switching to svn stable ?<br>
          </div>
          <div>(I usually prefer to update to new release versions when
            possible...) <br>
          </div>
          <div><br>
          </div>
          <div>Thanks,</div>
          <div>Marco</div>
        </div>
        <div><br>
        </div>
      </div>
      <div class="gmail_extra"><br>
        <div class="gmail_quote">2015-03-24 11:08 GMT+01:00 John Forrest
          <span dir="ltr">&lt;<a moz-do-not-send="true"
              href="mailto:john.forrest@fastercoin.com" target="_blank">john.forrest@fastercoin.com</a>&gt;</span>:<br>
          <blockquote class="gmail_quote" style="margin:0 0 0
            .8ex;border-left:1px #ccc solid;padding-left:1ex">
            <div bgcolor="#FFFFFF" text="#000000">
              <div>Marco,<br>
                <br>
                Thanks.  I wish I knew why Gomory cuts have started
                giving trouble.<br>
                <br>
                If I run with debug I get<br>
                <br>
                Cut with 7 coefficients, cuts off known solutions by
                1.403e-05, lo=-1.79769e+308, ub=9.75001<br>
                ( 3 , -9.75 ) ( 4 , -9.75 ) ( 6 , 9.75001 ) ( 7 ,
                9.75001 ) ( 8 , 9.75001 ) ( 9 , 9.75001 ) ( 10 , 9.75001
                ) <br>
                Non zero solution values are<br>
                ( 3 , 1 ) ( 4 , 1 ) ( 7 , 1 ) ( 9 , 1 ) ( 10 , 1 ) <br>
                <br>
                so you can see it is an accuracy problem.<br>
                <br>
                Yesterday I had changed stable so that CglGomory cuts
                were not default, but CglGMI was default - as CglGMI is
                more stable.  I then decided that was a bit drastic so
                changed it back unless a Cbc configure option is set.<br>
                <br>
                So in stable (svn) you now have several options.<br>
                <br>
                In configure use -DSWAP_GOMORY to use CglGMI as default
                (you can still switch back by using gomory ifmove etc.).<br>
                <br>
                In configure use -DWEAKEN_CUTS=1.  This weakens cuts a
                bit if they look slightly odd.<br>
                <br>
                If performance does not suffer, I would suggest first
                option - or just change to using -gomory off -gmi
                ifmove.<br>
                <br>
                John Forrest
                <div>
                  <div class="h5"><br>
                    <br>
                    On 24/03/15 08:23, mg wrote:<br>
                  </div>
                </div>
              </div>
              <blockquote type="cite">
                <div>
                  <div class="h5">
                    <div dir="ltr">Hi All,
                      <div>I'm currently solving some very small
                        multiknapsack-like problems and I got an
                        infeasibility on the attached problem (CBC
                        version release 2.9.3)</div>
                      <div><br>
                      </div>
                      <div>It seems related to some wrong cut, since
                        disabling the cut generators (with cutsonoff
                        off) it works.</div>
                      <div><br>
                      </div>
                      <div>I don't know if it's important, but if you
                        look at the attached LP, you can see a
                        constraint called "PrimaryObjAsConstr" that
                        comes from a previous solve where this
                        constraint was actually the objective function. </div>
                      <div>Well, the RHS of this contraint is the
                        objective value of the previous solve (4) plus a
                        tolerance (0.1). If I put it to 4, the model is
                        solved without any problem even with the cut
                        generators enabled. Does this lead to some
                        integer-related stuff ?</div>
                      <div><br>
                      </div>
                      <div>For the moment I can solve it by disabling
                        the cut generation, but I wanted to report the
                        issue that might rise again in other cases.</div>
                      <div><br>
                      </div>
                      <div>Thanks in advance,</div>
                      <div>Marco</div>
                    </div>
                    <br>
                    <fieldset></fieldset>
                    <br>
                  </div>
                </div>
                <pre>_______________________________________________
Cbc mailing list
<a moz-do-not-send="true" href="mailto:Cbc@list.coin-or.org" target="_blank">Cbc@list.coin-or.org</a>
<a moz-do-not-send="true" href="http://list.coin-or.org/mailman/listinfo/cbc" target="_blank">http://list.coin-or.org/mailman/listinfo/cbc</a>
</pre>
              </blockquote>
              <br>
            </div>
            <br>
            _______________________________________________<br>
            Cbc mailing list<br>
            <a moz-do-not-send="true" href="mailto:Cbc@list.coin-or.org">Cbc@list.coin-or.org</a><br>
            <a moz-do-not-send="true"
              href="http://list.coin-or.org/mailman/listinfo/cbc"
              target="_blank">http://list.coin-or.org/mailman/listinfo/cbc</a><br>
            <br>
          </blockquote>
        </div>
        <br>
      </div>
    </blockquote>
    <br>
  </body>
</html>