void printOptimal()
{
vectord sv(2);
sv(0) = 0.1238938; sv(1) = 0.818333;
std::cout << "Solutions: " << sv << "->"
<< evaluateSample(sv) << std::endl;
sv(0) = 0.5427728; sv(1) = 0.151667;
std::cout << "Solutions: " << sv << "->"
<< evaluateSample(sv) << std::endl;
sv(0) = 0.961652; sv(1) = 0.1650;
std::cout << "Solutions: " << sv << "->"
<< evaluateSample(sv) << std::endl;
}
double DiscreteModel::evaluateSampleInternal( const vectord &query )
{
const double yNext = evaluateSample(query);
if (yNext == HUGE_VAL)
{
throw std::runtime_error("Function evaluation out of range");
}
return yNext;
};
void DiscreteModel::sampleInitialPoints(matrixd& xPoints, vectord& yPoints)
{
vecOfvec perms = mInputSet;
// By using random permutations, we guarantee that
// the same point is not selected twice
utils::randomPerms(perms,mEngine);
// vectord xPoint(mInputSet[0].size());
for(size_t i = 0; i < yPoints.size(); i++)
{
const vectord xP = perms[i];
row(xPoints,i) = xP;
yPoints(i) = evaluateSample(xP);
}
}
/**
* \brief Wrapper for the target function normalize in the hypercube
* [0,1]
* @param query point to evaluate in [0,1] hypercube
* @return actual return value of the target function
*/
inline double evaluateSampleInternal( const vectord &query )
{
return evaluateSample(query);
};
/**
* \brief Wrapper for the target function normalize in the hypercube
* [0,1]
* @param query point to evaluate in [0,1] hypercube
* @return actual return value of the target function
*/
inline double evaluateSampleInternal( const vectord &query )
{
vectord unnormalizedQuery = mBB->unnormalizeVector(query);
return evaluateSample(unnormalizedQuery);
}; // evaluateSampleInternal
