Function GMP::Instance::CreatePresolved(GMP, name)

GMP::Instance::CreatePresolved

The function GMP::Instance::CreatePresolved generates a mathematical program that is the presolved representation of the specified generated mathematical program. The generated mathematical program can be a linear or nonlinear model, and should be generated using the function GMP::Instance::Generate.

GMP::Instance::CreatePresolved(
     GMP,            ! (input) a generated mathematical program
     name            ! (input) a string expression
)

Arguments

GMP

An element in the set AllGeneratedMathematicalPrograms.

name

A string that holds the name for the presolved mathematical program.

Return Value

A new element in the set AllGeneratedMathematicalPrograms, with the name as specified by the name argument, if the presolver did not find an infeasibility. Else, the empty element.

Note

• By using the functions GMP::Column::GetStatus and GMP::Row::GetStatus it is possible to check whether a column or row was deleted when the presolved mathematical program was created.

• By using the functions GMP::Column::GetLowerBound and GMP::Column::GetUpperBound it is possible to retrieve the lower and upper bound of a column in the presolved mathematical program.

• If the original GMP is deleted then the presolved GMP created by GMP::Instance::CreatePresolved will also be deleted.

• If the option MINLP Probing is switched on, then this function will change the mathematical programming type from MINLP (NLP) into MIP (LP) if the presolved model contains no nonlinear constraints.

• This function is not supported for GMP’s generated using GMP::Instance::CreateDual, GMP’s generated using GMP::Instance::GenerateRobustCounterpart, and GMP’s generated using GMP::Instance::GenerateStochasticProgram.

Example

Assume that MP is a mathematical program and gmpMP and gmpPre are element parameters with range AllGeneratedMathematicalPrograms. To solve the presolved model using GMP functions we can use:

gmpMP := GMP::Instance::Generate( MP );
gmpPre := GMP::Instance::CreatePresolved( gmpMP, "PresolvedModel" );

GMP::Instance::Solve( gmpPre ) ;

In case the GMP variant of the AOA module is used we can use:

gmpMP := GMP::Instance::Generate( MP );
gmpPre := GMP::Instance::CreatePresolved( gmpMP, "PresolvedModel" );

GMPOuterApprox::DoOuterApproximation( gmpPre );

Here GMPOuterApprox is the prefix used by the GMP Outer Approximation Module.

See also

The functions GMP::Instance::Delete, GMP::Instance::Generate, GMP::Instance::Solve, GMP::Column::GetStatus, GMP::Row::GetStatus, GMP::Column::GetLowerBound and GMP::Column::GetUpperBound.