[Couenne] A message showed "couenne: Infeasible problem"

[Pietro Belotti]

Dear Ote,

if you want Couenne to use Cplex, you should download the source code of 
Couenne and configure it using the two options below, specifying the 
appropriate paths for Cplex's libraries and include files. After that, 
follow the usual instructions for building the Couenne executable (you 
will need a compiler for Windows of for Cygwin).

Nevertheless, it seems that this is not the issue here. Your model seems 
infeasible, therefore whether or not you use Cplex won't change the 
result. If you need help understand why your model is infeasible, you may 
want to reply attaching the model you are trying to solve.

Hope this helps.

Best,
Pietro

--
Pietro Belotti
Dept. of Mathematical Sciences
Clemson University
email: pbelott at clemson.edu
phone: 864-656-6765
www:   myweb.clemson.edu/~pbelott

On Mon, 18 Oct 2010, Ote Bao wrote:

> Dear Sir,
> 
> After I ran my model on Couenne.   A message showed that
> 
> -----------------------------------------------------
> Error in an AMPL evaluation. Run with "halt_on_ampl_error yes" to see details.
> Exception of type: TOO_FEW_DOF in file "E:\stefan\nbBuildDir\Couenne\releases-0.3.2\Ipopt\src\Interfaces\IpIpoptApplication.cpp" at line 788:
>  Exception message: status != TOO_FEW_DEGREES_OF_FREEDOM evaluated false: Too few degrees of freedom (rethrown)!
> TODO: No statistics available from Ipopt in Bonmin::IpoptSolver::IterationCount
> TODO: No statistics available from Ipopt in Bonmin::IpoptSolver::CPUTime
> TODO: No statistics available from Ipopt in Bonmin::IpoptSolver::IterationCount
> TODO: No statistics available fro! m Ipopt in Bonmin::IpoptSolver::CPUTime
> TODO: No statistics available from Ipopt in Bonmin::IpoptSolver::IterationCount
> Exception of type: TOO_FEW_DOF in file "E:\stefan\nbBuildDir\Couenne\releases-0.3.2\Ipopt\src\Interfaces\IpIpoptApplication.cpp" at line 788:
>  Exception message: status != TOO_FEW_DEGREES_OF_FREEDOM evaluated false: Too few degrees of freedom (rethrown)!
> TODO: No statistics available from Ipopt in Bonmin::IpoptSolver::IterationCount
> TODO: No statistics available from Ipopt in Bonmin::IpoptSolver::CPUTime
> TODO: No statistics available from Ipopt in Bonmin::IpoptSolver::IterationCount
> TODO: No statistics available from Ipopt in Bonmin::IpoptSolver::CPUTime
> TODO: No statistics available from Ipopt in Bonmin::IpoptSolver::IterationCount
> Cbc0006I The LP relaxation is infeasible or too expensive
> Cbc0031I 4 added rows had average density of 2
> Cbc0013! I At root node, 4 cuts changed objective from 30 to 30 in 2 pa! ssesCbc0014I Cut generator 0 (Couenne convexifier cuts) - 4 row cuts average 2.0 elements, 34 column cuts (38 active)
> Cbc0001I Search completed - best objective 1e+050, took 4 iterations and 0 nodes (24.80 seconds)
> Cbc0035I Maximum depth 0, 0 variables fixed on reduced cost
>  Couenne convexifier cuts was tried 2 times and created 38 cuts of which 4 were active after adding rounds of cuts
>   "Finished"
> 
> couenne: Infeasible problem
> -----------------------------------------------------
> After I read options section in couenne: a user’s manual. 
> In 6.2(Linearization option) it said that...
> – lp_solver (clp): the Linear Programming solver used to find lower bounds.
> By default, CLP is used, but the user can also specify cplex to use the
> Cplex commercial software. In the latter case, the configure script has to
> be (re-)run with the following options:
> --with-cplex-lib="/your/path/to/libcplex.a -lpthread"
> --with-cplex-incdir=/your/path/to/include/ilcplex
> My computer platform is Window XP and also AMPL and CPLEX Version.11.1.0 installed.
> Could you please advise me?
> 
> Best Reg! ards, Ot e
> 
>