Re: [eclipse-clp-users] Trying to run MIP using eplex.

The help text for eplex:integers/1 states that

        [u]nlike integers/1 constraints from other solvers, the
        variables are not constrained to be integer type at the ECLiPSe
        level. However, when a typed_solution is retrieved (e.g. via
        eplex_var_get/3), this will be rounded to the nearest integer.

I.e., you have to retrieve the integer value from the MIP solver
yourself if you are not using optimize/2:

mip_example(X,Y,Cost) :-
     eplex_instance(my_instance),
     my_instance: (X+Y $>= 3),
     my_instance: (X-Y $= 0),
     my_instance: integers([X]),  % This is a mixed integer problem.
     my_instance: eplex_solver_setup(min(X)),
     my_instance: eplex_solve(Cost),
     % retrieve typed solution
     my_instance: eplex_get(vars, Vars),
     my_instance: eplex_get(typed_solution, Vals),
     Vars = Vals.

?- mip_example(X, Y, C).
X = 2
Y = 2.0
C = 2.0
Yes (0.02s cpu)


Cheers,
Thorsten


Am 10.10.2013 17:04, schrieb Paul Cannongeyser:
> I ran this example.  The results were unexpected.
>
> ----- eplex_9.ecl
> % Run in this way: mip_example(X,Y,Cost).
> % From http://eclipseclp.org/doc/libman/libman060.html#toc67 entitled
> 10.2.6  Examples
> /*
> Results:
> ?- mip_example(X, Y, Cost).
> X = X{-1.7976931348623157e+308 .. 1.7976931348623157e+308 @ 2.0}
> Y = Y{-1.7976931348623157e+308 .. 1.7976931348623157e+308 @ 2.0}
> Cost = 2.0
> Yes (0.00s cpu)
>
> No integer variables - nothing to do (in orange letters)
>
> */
>
> :- lib(eplex).
> :- eplex_instance(my_instance).
>
> mip_example(X,Y,Cost) :-
>      my_instance: (X+Y $>= 3),
>      my_instance: (X-Y $= 0),
>      my_instance: integers([X]),  % This is a mixed integer problem.
>      my_instance: eplex_solver_setup(min(X)),
>      my_instance: eplex_solve(Cost).
> -----
>
> However, I was able to run a similar example without any warnings and
> the result showed as an integer, rather 
> than a decimal.
>
> ----- eplex_3.ecl
> % Run in this way: top(X,Y,Cost).
> % From http://eclipseclp.org/reports/Workshop/eplex.ppt entitled The
> EPLEX Library
> /*
> Results:
> ?- top(X, Y, Cost).
> X = 2               <<<<<<====== X is printed as an integer, 2, not a
> decimal, 2.0.  (Nicely done.)
> Y = 2.0
> Cost = 2.0
> Yes (0.00s cpu)
>
> */
>
> :-lib(eplex).
>
> top(X,Y,Cost) :-
> X+Y $>=3,
> X-Y $= 0,
> integers([X]),  % This is a mixed integer problem.
> optimize(min(X),Cost).
> -----
>
> Is the result in my first program run (i.e., eplex_9.ecl, at the top
> of this email) the correct behavior or did I do something wrong?
>
>
> ------------------------------------------------------------------------------
> October Webinars: Code for Performance
> Free Intel webinars can help you accelerate application performance.
> Explore tips for MPI, OpenMP, advanced profiling, and more. Get the most from 
> the latest Intel processors and coprocessors. See abstracts and register >
> http://pubads.g.doubleclick.net/gampad/clk?id=60134071&iu=/4140/ostg.clktrk
>
>
> _______________________________________________
> ECLiPSe-CLP-Users mailing list
> ECLiPSe-CLP-Users_at_lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/eclipse-clp-users