Changeset 14

Nov 4, 2008 12:03:57 AM (12 years ago)

added couenne.opt, will fill it soon

2 edited


  • trunk/Couenne/src/couenne.opt

    r13 r14  
     2display statistics at the end of the run
     5Output level for braching code in Couenne
     8Output level for bound tightening code in Couenne
     11Output level for convexifying code in Couenne
     14Output level for problem manipulation code in Couenne
     17Output level for NLP heuristic in Couenne
     20Output level for disjunctive cuts in Couenne
     23Multipliers of pseudocosts for estimating and update estimation of bound
     26Use distance between LP points to update multipliers of pseudocosts " 
     29Use Special Ordered Sets (SOS) as indicated in the MINLP model
     32Apply bound tightening before branching
     35Apply convexification cuts before branching (for now only within strong branching)
     38Chooses branching point selection strategy
     41Defines convex combination of mid point and current LP point: "
     44Defines safe interval percentage for using LP point as a branching point
     47Priority of continuous variable branching
     50Apply Reduced Cost Branching (instead of the Violation Transfer) -- MUST have vt_obj enabled
     53Specify the frequency (in terms of nodes) at which couenne ecp cuts are generated.
     56Check all LPs through an independent call to OsiClpSolverInterface::initialSolve()
     59Do we search for local solutions of NLP's
     62Specify the logarithm of the number of local optimizations to perform"
     65Deterimnes in which point the linear over/under-estimator are generated
     68Specify the number of points at which to convexify when convexification type"
     71Yes if only violated convexification cuts should be added
     74Artificial cutoff
     77Window around known optimum
     80Tolerance for constraints/auxiliary variables
     83Feasibility-based (cheap) bound tightening
     86Use quadratic expressions and related exprQuad class
     89Optimality-based (expensive) bound tightening
     92Specify the frequency (in terms of nodes) for optimality-based bound tightening.
     95Aggressive feasibility-based bound tightening (to use with NLP points)
     98Specify the frequency (in terms of nodes) for aggressive bound tightening.
     101Artificial lower bound
     104type of branching object for variable selection
  • trunk/INSTALL

    r11 r14  
    127127the "g" after the "write" keyword).
     129Couenne can be fine tuned by setting parameters in the option file couenne.opt,
     130which is read from the same directory where Couenne is launched. A typical
     131couenne.opt is provided in Couenne/src/, with some explanation on how to use
     132its parameters.
Note: See TracChangeset for help on using the changeset viewer.