<html><head><meta http-equiv="Content-Type" content="text/html; charset=utf-8"></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; line-break: after-white-space;" class=""><span style="caret-color: rgb(0, 0, 0); color: rgb(0, 0, 0);" class="">Hello Mr. Bellotti</span><span style="caret-color: rgb(0, 0, 0); color: rgb(0, 0, 0);" class="">,</span><div class="" style="caret-color: rgb(0, 0, 0); color: rgb(0, 0, 0);"><br class=""></div><div class="" style="caret-color: rgb(0, 0, 0); color: rgb(0, 0, 0);">I’m a master student in Boğaziçi University, Istanbul-Turkey. I am currently working in a project for the next upgrade will take in place at CERN and I am trying to optimise some aspects of the new detector that is planning to be build. For this I’m using Couenne solver and in short our model of the problem has the goal of minimising the standard deviation of the found distribution, thus our problem is a non-convex nonlinear and the components that are needed to be distributed are integers (binaries to be exact), thus Couenne solver seemed like a good option to our problem. </div><div class="" style="caret-color: rgb(0, 0, 0); color: rgb(0, 0, 0);"><br class=""></div><div class="" style="caret-color: rgb(0, 0, 0); color: rgb(0, 0, 0);">We can separate our data into small groups for testing the model and so far we were working on a set that consists of ˜900 variables and everything seems working fine. The problem arise when we try to increase this number to its multiplications. It can be thought like 900 variables is for one sector and we have in total 28 sectors. Naturally, combination that Couenne needs to solve increases exponentially and that’s why we were trying to find good options to reduce the cost of calculations that are needed to solve the problem. One candidate was CBC solver options since they are basically manipulates the tree and hence the number of trial points. However, we could not make it work any option for CBC in Couenne so far. </div><div class="" style="caret-color: rgb(0, 0, 0); color: rgb(0, 0, 0);"><br class=""></div><div class="" style="caret-color: rgb(0, 0, 0); color: rgb(0, 0, 0);">The way I am trying is like it is explained in the user manual, as in the “couenne.opt’ file I put the lines:</div><div class=""><div class=""><font color="#000000" class=""><span style="caret-color: rgb(0, 0, 0);" class="">time_limit 500 </span></font></div><div class=""><font color="#000000" class=""><span style="caret-color: rgb(0, 0, 0);" class="">aggressive_fbbt no</span></font></div><div class=""><font color="#000000" class=""><span style="caret-color: rgb(0, 0, 0);" class="">optimality_bt no</span></font></div><div class=""><font color="#000000" class=""><span style="caret-color: rgb(0, 0, 0);" class="">cbc.nodlim 15000</span></font></div><div class=""><font color="#000000" class=""><span style="caret-color: rgb(0, 0, 0);" class=""><br class=""></span></font></div><div class=""><font color="#000000" class="">But CBC option raises an error of the form: </font></div><div class=""><div class="lm-Widget p-Widget p-Panel jp-OutputArea-child lm-Panel"><div class="lm-Widget p-Widget jp-RenderedText jp-mod-trusted jp-OutputArea-output" data-mime-type="application/vnd.jupyter.stderr"><pre class="">libc++abi.dylib: terminating with uncaught exception of type Ipopt::IpoptException
</pre></div></div><div class="lm-Widget p-Widget p-Panel jp-OutputArea-child lm-Panel"></div><div class="lm-Widget p-Widget jp-RenderedText jp-mod-trusted jp-OutputArea-output" data-mime-type="application/vnd.jupyter.stdout"><pre class="">Couenne 0.5.7 -- an Open-Source solver for Mixed Integer Nonlinear Optimization
Mailing list: <a href="mailto:couenne@list.coin-or.org" class="">couenne@list.coin-or.org</a>
Instructions: <a href="http://www.coin-or.org/COUENNE" rel="noopener" target="_blank" class="">http://www.coin-or.org/Couenne</a>
Exception of type: OPTION_INVALID in file "/Users/runner/runners/2.160.1/work/1/s/Ipopt/Ipopt/src/Common/IpOptionsList.cpp" at line 639:
Exception message: Read Option: "cbc.nodlim". It is not a valid option. Check the list of available options.
ERROR: Solver (asl) returned non-zero return code (-6)
ERROR: See the solver log above for diagnostic information.
</pre></div><div class="">I am using binary file of Couenne for macOS that I downloaded from AMPL Open Source Solvers and I am using it in Pyomo (version 5.7.1) platform.</div><div class=""><br class=""></div><div class="">My question is two folds; </div><div class="">First, how can I use CBC options in Couenne through or not couenne.opt file if it is possible. </div><div class="">And secondly, I would like to ask you if you can give us some advices about dealing with this many of variables effectively.</div></div></div><div class="" style="caret-color: rgb(0, 0, 0); color: rgb(0, 0, 0);"><br class=""></div><div class="" style="caret-color: rgb(0, 0, 0); color: rgb(0, 0, 0);">Thank you so much in advance,</div><div class="" style="caret-color: rgb(0, 0, 0); color: rgb(0, 0, 0);">My kindest regards.</div><div class="" style="caret-color: rgb(0, 0, 0); color: rgb(0, 0, 0);">Yusuf Can Cekmecelioglu</div></body></html>