TEST_F(AnalysisDriverFixture,RuntimeBehavior_StopCustomAnalysis) {
// Tests for stopping time < 20s.
// RETRIEVE PROBLEM
Problem problem = retrieveProblem("UserScriptContinuous",true,false);
// DEFINE SEED
Model model = model::exampleModel();
openstudio::path p = toPath("./example.osm");
model.save(p,true);
FileReference seedModel(p);
// CREATE ANALYSIS
Analysis analysis("Stop Custom Analysis",
problem,
seedModel);
// generate 100 random points
boost::mt19937 mt;
typedef boost::uniform_real<> dist_type;
typedef boost::variate_generator<boost::mt19937&, dist_type > gen_type;
InputVariableVector variables = problem.variables();
ContinuousVariable cvar = variables[0].cast<ContinuousVariable>();
gen_type generator0(mt,dist_type(cvar.minimum().get(),cvar.maximum().get()));
cvar = variables[1].cast<ContinuousVariable>();
gen_type generator1(mt,dist_type(cvar.minimum().get(),cvar.maximum().get()));
cvar = variables[2].cast<ContinuousVariable>();
gen_type generator2(mt,dist_type(cvar.minimum().get(),cvar.maximum().get()));
for (int i = 0, n = 100; i < n; ++i) {
std::vector<QVariant> values;
double value = generator0();
values.push_back(value);
value = generator1();
values.push_back(value);
value = generator2();
values.push_back(value);
OptionalDataPoint dataPoint = problem.createDataPoint(values);
ASSERT_TRUE(dataPoint);
ASSERT_TRUE(analysis.addDataPoint(*dataPoint));
}
// RUN ANALYSIS
ProjectDatabase database = getCleanDatabase("StopCustomAnalysis");
AnalysisDriver analysisDriver(database);
AnalysisRunOptions runOptions = standardRunOptions(analysisDriver.database().path().parent_path());
runOptions.setQueueSize(2);
StopWatcher watcher(analysisDriver);
watcher.watch(analysis.uuid());
CurrentAnalysis currentAnalysis = analysisDriver.run(analysis,runOptions);
EXPECT_EQ(2,currentAnalysis.numQueuedJobs());
EXPECT_EQ(0,currentAnalysis.numQueuedDakotaJobs());
EXPECT_EQ(100,currentAnalysis.totalNumJobsInOSIteration());
EXPECT_EQ(0,currentAnalysis.numCompletedJobsInOSIteration());
analysisDriver.waitForFinished();
EXPECT_FALSE(analysisDriver.isRunning());
EXPECT_GE(watcher.nComplete(),watcher.stopNum());
EXPECT_LE(watcher.stoppingTime(),openstudio::Time(0,0,0,20));
// check conditions afterward
RunManager runManager = analysisDriver.database().runManager();
EXPECT_FALSE(runManager.workPending());
BOOST_FOREACH(const Job& job,runManager.getJobs()) {
EXPECT_FALSE(job.running());
EXPECT_FALSE(job.treeRunning());
}
EXPECT_TRUE(currentAnalysis.numCompletedJobsInOSIteration() > 0);
EXPECT_TRUE(currentAnalysis.analysis().dataPointsToQueue().size() > 0u);
EXPECT_TRUE(currentAnalysis.analysis().dataPointsToQueue().size() < 100u);
EXPECT_EQ(0u,analysisDriver.currentAnalyses().size());
}
/*
* BIG MACHINE
*/
std::list<LabelPosition*>* Pal::labeller( int nbLayers, char **layersName , double *layersFactor, double scale, double bbox[4], PalStat **stats, bool displayAll )
{
#ifdef _DEBUG_
std::cout << "LABELLER (selection)" << std::endl;
#endif
Problem *prob;
SearchMethod old_searchMethod = searchMethod;
if ( displayAll )
{
setSearch( POPMUSIC_TABU );
}
#ifdef _VERBOSE_
clock_t start = clock();
double create_time;
std::cout << std::endl << "bbox: " << bbox[0] << " " << bbox[1] << " " << bbox[2] << " " << bbox[3] << std::endl;
#endif
#ifdef _EXPORT_MAP_
// TODO this is not secure
std::ofstream svgmap( "pal-map.svg" );
svgmap << "<?xml version=\"1.0\" encoding=\"UTF-8\" standalone=\"no\"?>" << std::endl
<< "<svg" << std::endl
<< "xmlns:dc=\"http://purl.org/dc/elements/1.1/\"" << std::endl
<< "xmlns:cc=\"http://creativecommons.org/ns#\"" << std::endl
<< "xmlns:rdf=\"http://www.w3.org/1999/02/22-rdf-syntax-ns#\"" << std::endl
<< "xmlns:svg=\"http://www.w3.org/2000/svg\"" << std::endl
<< "xmlns=\"http://www.w3.org/2000/svg\"" << std::endl
<< "xmlns:sodipodi=\"http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd\"" << std::endl
<< "xmlns:inkscape=\"http://www.inkscape.org/namespaces/inkscape\"" << std::endl
<< "width=\"" << convert2pt( bbox[2] - bbox[0], scale, dpi ) << "\"" << std::endl
<< "height=\"" << convert2pt( bbox[3] - bbox[1], scale, dpi ) << "\">" << std::endl; // TODO xmax ymax
#endif
QTime t;
t.start();
// First, extract the problem
// TODO which is the minimum scale ? (> 0, >= 0, >= 1, >1 )
if ( scale < 1 || ( prob = extract( nbLayers, layersName, layersFactor, bbox[0], bbox[1], bbox[2], bbox[3], scale,
#ifdef _EXPORT_MAP_
& svgmap
#else
NULL
#endif
) ) == NULL )
{
#ifdef _VERBOSE_
if ( scale < 1 )
std::cout << "Scale is 1:" << scale << std::endl;
else
std::cout << "empty problem... finishing" << std::endl;
#endif
#ifdef _EXPORT_MAP_
svgmap << "</svg>" << std::endl;
svgmap.close();
#endif
// nothing to be done => return an empty result set
if ( stats )
( *stats ) = new PalStat();
return new std::list<LabelPosition*>();
}
std::cout << "PAL EXTRACT: " << t.elapsed() / 1000.0 << " s" << std::endl;
t.restart();
// reduce number of candidates
// (remove candidates which surely won't be used)
prob->reduce();
#ifdef _VERBOSE_
std::cerr << prob->nblp << "\t"
<< prob->nbOverlap;
#endif
prob->displayAll = displayAll;
#ifdef _VERBOSE_
create_time = double( clock() - start ) / double( CLOCKS_PER_SEC );
std::cout << std::endl << "Problem : " << prob->nblp << " candidates for " << prob->nbft << " features makes " << prob->nbOverlap << " overlaps" << std::endl;
std::cout << std::endl << "Times:" << std::endl << " to create problem: " << create_time << std::endl;
#endif
// search a solution
if ( searchMethod == FALP )
prob->init_sol_falp();
else if ( searchMethod == CHAIN )
prob->chain_search();
else
prob->popmusic();
std::cout << "PAL SEARCH (" << searchMethod << "): " << t.elapsed() / 1000.0 << " s" << std::endl;
t.restart();
// Post-Optimization
//prob->post_optimization();
std::list<LabelPosition*> * solution = prob->getSolution( displayAll );
if ( stats )
*stats = prob->getStats();
#ifdef _EXPORT_MAP_
prob->drawLabels( svgmap );
svgmap << "</svg>" << std::endl;
svgmap.close();
#endif
#ifdef _VERBOSE_
clock_t total_time = clock() - start;
std::cout << " Total time: " << double( total_time ) / double( CLOCKS_PER_SEC ) << std::endl;
std::cerr << "\t" << create_time << "\t" << double( total_time ) / double( CLOCKS_PER_SEC ) << std::endl;
#endif
delete prob;
if ( displayAll )
{
setSearch( old_searchMethod );
}
return solution;
}
bool solve_translations_problem_l2_chordal
(
const int* edges,
const double* poses,
const double* weights,
int num_edges,
double loss_width,
double* X,
double function_tolerance,
double parameter_tolerance,
int max_iterations
)
{
// seed the random number generator
std::srand( std::time( nullptr ) );
// re index the edges to be a sequential set
std::vector<int> reindexed_edges(edges, edges+2*num_edges);
std::vector<int> reindexed_lookup;
reindex_problem(&reindexed_edges[0], num_edges, reindexed_lookup);
const int num_nodes = reindexed_lookup.size();
// Init with a random guess solution
std::vector<double> x(3*num_nodes);
for (int i=0; i<3*num_nodes; ++i)
x[i] = (double)rand() / RAND_MAX;
// add the parameter blocks (a 3-vector for each node)
Problem problem;
for (int i=0; i<num_nodes; ++i)
problem.AddParameterBlock(&x[3*i], 3);
// set the residual function (chordal distance for each edge)
for (int i=0; i<num_edges; ++i) {
CostFunction* cost_function =
new AutoDiffCostFunction<ChordFunctor, 3, 3, 3>(
new ChordFunctor(poses+3*i, weights[i]));
if (loss_width == 0.0) {
// No robust loss function
problem.AddResidualBlock(cost_function, nullptr, &x[3*reindexed_edges[2*i+0]], &x[3*reindexed_edges[2*i+1]]);
} else {
problem.AddResidualBlock(cost_function, new ceres::HuberLoss(loss_width), &x[3*reindexed_edges[2*i+0]], &x[3*reindexed_edges[2*i+1]]);
}
}
// Fix first camera in {0,0,0}: fix the translation ambiguity
x[0] = x[1] = x[2] = 0.0;
problem.SetParameterBlockConstant(&x[0]);
// solve
Solver::Options options;
#ifdef OPENMVG_USE_OPENMP
options.num_threads = omp_get_max_threads();
#if CERES_VERSION_MAJOR < 2
options.num_linear_solver_threads = omp_get_max_threads();
#endif
#endif // OPENMVG_USE_OPENMP
options.minimizer_progress_to_stdout = false;
options.logging_type = ceres::SILENT;
options.max_num_iterations = max_iterations;
options.function_tolerance = function_tolerance;
options.parameter_tolerance = parameter_tolerance;
// Since the problem is sparse, use a sparse solver iff available
if (ceres::IsSparseLinearAlgebraLibraryTypeAvailable(ceres::SUITE_SPARSE))
{
options.sparse_linear_algebra_library_type = ceres::SUITE_SPARSE;
options.linear_solver_type = ceres::SPARSE_NORMAL_CHOLESKY;
}
else if (ceres::IsSparseLinearAlgebraLibraryTypeAvailable(ceres::CX_SPARSE))
{
options.sparse_linear_algebra_library_type = ceres::CX_SPARSE;
options.linear_solver_type = ceres::SPARSE_NORMAL_CHOLESKY;
}
else if (ceres::IsSparseLinearAlgebraLibraryTypeAvailable(ceres::EIGEN_SPARSE))
{
options.sparse_linear_algebra_library_type = ceres::EIGEN_SPARSE;
options.linear_solver_type = ceres::SPARSE_NORMAL_CHOLESKY;
}
else
{
options.linear_solver_type = ceres::DENSE_NORMAL_CHOLESKY;
}
Solver::Summary summary;
Solve(options, &problem, &summary);
std::cout << summary.FullReport() << "\n";
if (summary.IsSolutionUsable())
{
// undo the re indexing
for (int i=0; i<num_nodes; ++i) {
const int j = reindexed_lookup[i];
X[3*j+0] = x[3*i+0];
X[3*j+1] = x[3*i+1];
X[3*j+2] = x[3*i+2];
}
}
return summary.IsSolutionUsable();
}
EndCriteria::Type
LineSearchBasedMethod::minimize(Problem& P,
const EndCriteria& endCriteria) {
// Initializations
Real ftol = endCriteria.functionEpsilon();
Size maxStationaryStateIterations_
= endCriteria.maxStationaryStateIterations();
EndCriteria::Type ecType = EndCriteria::None; // reset end criteria
P.reset(); // reset problem
Array x_ = P.currentValue(); // store the starting point
Size iterationNumber_ = 0;
// dimension line search
lineSearch_->searchDirection() = Array(x_.size());
bool done = false;
// function and squared norm of gradient values;
Real fnew, fold, gold2;
Real fdiff;
// classical initial value for line-search step
Real t = 1.0;
// Set gradient g at the size of the optimization problem
// search direction
Size sz = lineSearch_->searchDirection().size();
Array prevGradient(sz), d(sz), sddiff(sz), direction(sz);
// Initialize cost function, gradient prevGradient and search direction
P.setFunctionValue(P.valueAndGradient(prevGradient, x_));
P.setGradientNormValue(DotProduct(prevGradient, prevGradient));
lineSearch_->searchDirection() = -prevGradient;
bool first_time = true;
// Loop over iterations
do {
// Linesearch
if (!first_time)
prevGradient = lineSearch_->lastGradient();
t = (*lineSearch_)(P, ecType, endCriteria, t);
// don't throw: it can fail just because maxIterations exceeded
//QL_REQUIRE(lineSearch_->succeed(), "line-search failed!");
if (lineSearch_->succeed())
{
// Updates
// New point
x_ = lineSearch_->lastX();
// New function value
fold = P.functionValue();
P.setFunctionValue(lineSearch_->lastFunctionValue());
// New gradient and search direction vectors
// orthogonalization coef
gold2 = P.gradientNormValue();
P.setGradientNormValue(lineSearch_->lastGradientNorm2());
// conjugate gradient search direction
direction = getUpdatedDirection(P, gold2, prevGradient);
sddiff = direction - lineSearch_->searchDirection();
lineSearch_->searchDirection() = direction;
// Now compute accuracy and check end criteria
// Numerical Recipes exit strategy on fx (see NR in C++, p.423)
fnew = P.functionValue();
fdiff = 2.0*std::fabs(fnew-fold) /
(std::fabs(fnew) + std::fabs(fold) + QL_EPSILON);
if (fdiff < ftol ||
endCriteria.checkMaxIterations(iterationNumber_, ecType)) {
endCriteria.checkStationaryFunctionValue(0.0, 0.0,
maxStationaryStateIterations_, ecType);
endCriteria.checkMaxIterations(iterationNumber_, ecType);
return ecType;
}
P.setCurrentValue(x_); // update problem current value
++iterationNumber_; // Increase iteration number
first_time = false;
} else {
done = true;
}
} while (!done);
P.setCurrentValue(x_);
return ecType;
}
EndCriteria::Type DifferentialEvolution::minimize(Problem& P,
const EndCriteria& endCriteria) {
EndCriteria::Type ecType = EndCriteria::MaxIterations;
QL_REQUIRE(P.currentValue().size() == nParam_,
"Number of parameters mismatch between problem and DE optimizer");
P.reset();
init();
Real bestCost = QL_MAX_REAL;
Size bestPop = 0;
for (Size p = 0; p < nPop_; ++p) {
Array tmp(currGen_[p].pop_);
try {
currGen_[p].cost_ = P.costFunction().value(tmp);
} catch (Error&) {
currGen_[p].cost_ = QL_MAX_REAL;
}
if (currGen_[p].cost_ < bestCost) {
bestPop = p;
bestCost = currGen_[p].cost_;
}
}
Size lastChange = 0;
Size lastParamChange = 0;
for(Size i=0; i<endCriteria.maxIterations(); ++i) {
Size newBestPop = bestPop;
Real newBestCost = bestCost;
for (Size p=0; p<nPop_; ++p) {
// Find 3 different populations randomly
Size r1;
do {
r1 = static_cast <Size> (uniformRng_.nextInt32() % nPop_);
}
while(r1 == p || r1 == bestPop);
Size r2;
do {
r2 = static_cast <Size> (uniformRng_.nextInt32() % nPop_);
}
while ( r2 == p || r2 == bestPop || r2 == r1);
Size r3;
do {
r3 = static_cast <Size> (uniformRng_.nextInt32() % nPop_);
} while ( r3 == p || r3 == bestPop || r3 == r1 || r3 == r2);
for(Size j=0; j<nParam_; ++j) {
nextGen_[p].pop_[j] = currGen_[p].pop_[j];
}
Size j = static_cast <Size> (uniformRng_.nextInt32() % nParam_);
Size L = 0;
do {
const double tmp =
currGen_[ p].pop_[j] * a0_
+ currGen_[ r1].pop_[j] * a1_
+ currGen_[ r2].pop_[j] * a2_
+ currGen_[ r3].pop_[j] * a3_
+ currGen_[bestPop].pop_[j] * aBest_;
nextGen_[p].pop_[j] =
std::min(maxParams_[j], std::max(minParams_[j], tmp));
j = (j+1)%nParam_;
++L;
} while ((uniformRng_.nextReal() < CR_) && (L < nParam_));
// Evaluate the new population
Array tmp(nextGen_[p].pop_);
try {
nextGen_[p].cost_ = P.costFunction().value(tmp);
} catch (Error&) {
nextGen_[p].cost_ = QL_MAX_REAL;
}
// Not better, discard it and keep the old one.
if (nextGen_[p].cost_ >= currGen_[p].cost_) {
nextGen_[p] = currGen_[p];
}
// Better, keep it.
else {
// New best?
if (nextGen_[p].cost_ < newBestCost) {
newBestPop = p;
newBestCost = nextGen_[p].cost_;
}
}
}
if(std::abs(newBestCost-bestCost) > endCriteria.functionEpsilon()) {
lastChange = i;
}
const Array absDiff = Abs(nextGen_[newBestPop].pop_-currGen_[bestPop].pop_);
if(*std::max_element(absDiff.begin(), absDiff.end()) > endCriteria.rootEpsilon()) {
lastParamChange = i;
}
bestPop = newBestPop;
bestCost = newBestCost;
currGen_ = nextGen_;
if(i-lastChange > endCriteria.maxStationaryStateIterations()) {
ecType = EndCriteria::StationaryFunctionValue;
break;
}
if(i-lastParamChange > endCriteria.maxStationaryStateIterations()) {
ecType = EndCriteria::StationaryPoint;
break;
}
if (adaptive_) adaptParameters();
}
const Array res(currGen_[bestPop].pop_);
P.setCurrentValue(res);
P.setFunctionValue(bestCost);
return ecType;
}
void TDefense::problemAction()
{
if (myProblems.size() > 0) {
for (int i = 0; i < myProblems.size(); i++)
{
Problem *curProblem = myProblems[i];
curProblem->setVX(0);
curProblem->setVY(1);
double x = curProblem->getX();
double y = curProblem->getY();
double vx = curProblem->getVX();
double vy = curProblem->getVY();
int height = curProblem->getImage().height();
int width = curProblem->getImage().width();
if ((y <= 0 && vy < 0) || (x <= 0 && vx < 0 || (gpHeight <= (y + height)) && y > 0 && vy > 0) || ((gpWidth == (x + width)) && vx > 0 && vx > 0))
{
if (!curProblem->isDropped)
{
numEnemiesDropped++;
extensions--;
IQ--;
curProblem->isDropped = true;
}
}
curProblem->move();
curProblem->setVX(vx);
curProblem->setVY(vy);
} //end outer for
}
}
EndCriteria::Type LevenbergMarquardt::minimize(Problem& P,
const EndCriteria& endCriteria) {
EndCriteria::Type ecType = EndCriteria::None;
P.reset();
Array x_ = P.currentValue();
currentProblem_ = &P;
initCostValues_ = P.costFunction().values(x_);
int m = initCostValues_.size();
int n = x_.size();
boost::scoped_array<double> xx(new double[n]);
std::copy(x_.begin(), x_.end(), xx.get());
boost::scoped_array<double> fvec(new double[m]);
boost::scoped_array<double> diag(new double[n]);
int mode = 1;
double factor = 1;
int nprint = 0;
int info = 0;
int nfev =0;
boost::scoped_array<double> fjac(new double[m*n]);
int ldfjac = m;
boost::scoped_array<int> ipvt(new int[n]);
boost::scoped_array<double> qtf(new double[n]);
boost::scoped_array<double> wa1(new double[n]);
boost::scoped_array<double> wa2(new double[n]);
boost::scoped_array<double> wa3(new double[n]);
boost::scoped_array<double> wa4(new double[m]);
// requirements; check here to get more detailed error messages.
QL_REQUIRE(n > 0, "no variables given");
QL_REQUIRE(m >= n,
"less functions (" << m <<
") than available variables (" << n << ")");
QL_REQUIRE(endCriteria.functionEpsilon() >= 0.0,
"negative f tolerance");
QL_REQUIRE(xtol_ >= 0.0, "negative x tolerance");
QL_REQUIRE(gtol_ >= 0.0, "negative g tolerance");
QL_REQUIRE(endCriteria.maxIterations() > 0,
"null number of evaluations");
// call lmdif to minimize the sum of the squares of m functions
// in n variables by the Levenberg-Marquardt algorithm.
MINPACK::LmdifCostFunction lmdifCostFunction =
boost::bind(&LevenbergMarquardt::fcn, this, _1, _2, _3, _4, _5);
MINPACK::lmdif(m, n, xx.get(), fvec.get(),
static_cast<double>(endCriteria.functionEpsilon()),
static_cast<double>(xtol_),
static_cast<double>(gtol_),
static_cast<int>(endCriteria.maxIterations()),
static_cast<double>(epsfcn_),
diag.get(), mode, factor,
nprint, &info, &nfev, fjac.get(),
ldfjac, ipvt.get(), qtf.get(),
wa1.get(), wa2.get(), wa3.get(), wa4.get(),
lmdifCostFunction);
info_ = info;
// check requirements & endCriteria evaluation
QL_REQUIRE(info != 0, "MINPACK: improper input parameters");
//QL_REQUIRE(info != 6, "MINPACK: ftol is too small. no further "
// "reduction in the sum of squares "
// "is possible.");
if (info != 6) ecType = QuantLib::EndCriteria::StationaryFunctionValue;
//QL_REQUIRE(info != 5, "MINPACK: number of calls to fcn has "
// "reached or exceeded maxfev.");
endCriteria.checkMaxIterations(nfev, ecType);
QL_REQUIRE(info != 7, "MINPACK: xtol is too small. no further "
"improvement in the approximate "
"solution x is possible.");
QL_REQUIRE(info != 8, "MINPACK: gtol is too small. fvec is "
"orthogonal to the columns of the "
"jacobian to machine precision.");
// set problem
std::copy(xx.get(), xx.get()+n, x_.begin());
P.setCurrentValue(x_);
P.setFunctionValue(P.costFunction().value(x_));
return ecType;
}
TEST_F(ProjectFixture,AnalysisRecord_AddAndRemoveDataPoints) {
// create an analysis with data points, no results
Analysis analysis("My Analysis",
Problem("My Problem",VariableVector(),runmanager::Workflow()),
FileReferenceType::OSM);
Problem problem = analysis.problem();
DiscretePerturbationVector perturbations;
int pi = 1;
std::stringstream ss;
for (int i = 0; i < 3; ++i) {
perturbations.push_back(NullPerturbation());
for (int j = 0; j < 4; ++j) {
ss << "measure" << pi << ".rb";
perturbations.push_back(RubyPerturbation(toPath(ss.str()),
FileReferenceType::OSM,
FileReferenceType::OSM,true));
ss.str("");
++pi;
}
ss << "Variable " << i+1;
problem.push(DiscreteVariable(ss.str(),perturbations));
perturbations.clear();
ss.str("");
}
EXPECT_EQ(3u,analysis.problem().variables().size());
ASSERT_FALSE(problem.variables().empty());
EXPECT_EQ(5u,problem.variables()[0].cast<DiscreteVariable>().numPerturbations(true));
problem.pushResponse(
LinearFunction("Energy Use",
VariableVector(1u,OutputAttributeVariable("Energy Use","Total.Energy.Use"))));
EXPECT_EQ(1u,problem.responses().size());
std::vector<QVariant> values(3u,0);
OptionalDataPoint dataPoint = problem.createDataPoint(values);
ASSERT_TRUE(dataPoint);
bool test = analysis.addDataPoint(*dataPoint);
EXPECT_TRUE(test);
values[1] = 3;
dataPoint = problem.createDataPoint(values);
ASSERT_TRUE(dataPoint);
test = analysis.addDataPoint(*dataPoint);
EXPECT_TRUE(test);
values[0] = 1;
values[1] = 2;
values[2] = 4;
dataPoint = problem.createDataPoint(values);
ASSERT_TRUE(dataPoint);
test = analysis.addDataPoint(*dataPoint);
EXPECT_TRUE(test);
EXPECT_TRUE(analysis.isDirty());
EXPECT_TRUE(problem.isDirty());
BOOST_FOREACH(const Variable& variable, problem.variables()) {
EXPECT_TRUE(variable.isDirty());
ASSERT_TRUE(variable.optionalCast<DiscreteVariable>());
BOOST_FOREACH(const DiscretePerturbation& perturbation,variable.cast<DiscreteVariable>().perturbations(false)) {
EXPECT_TRUE(perturbation.isDirty());
}
}
BOOST_FOREACH(const Function& response, problem.responses()) {
EXPECT_TRUE(response.isDirty());
}
BOOST_FOREACH(const DataPoint& dataPoint, analysis.dataPoints()) {
EXPECT_TRUE(dataPoint.isDirty());
}
// save to database, make sure everything is there, make sure clean signal filters down
ProjectDatabase database = getCleanDatabase("AnalysisRecord_AddAndRemoveDataPoints");
{
bool transactionStarted = database.startTransaction();
EXPECT_TRUE(transactionStarted);
AnalysisRecord analysisRecord(analysis,database);
database.save();
test = database.commitTransaction();
EXPECT_TRUE(test);
EXPECT_EQ(analysis.uuid(),analysisRecord.handle());
EXPECT_EQ(analysis.versionUUID(),analysisRecord.uuidLast());
ASSERT_NO_THROW(analysisRecord.problemRecord());
ProblemRecord problemRecord = analysisRecord.problemRecord();
InputVariableRecordVector variableRecords = problemRecord.inputVariableRecords();
EXPECT_EQ(3u,variableRecords.size());
ASSERT_FALSE(variableRecords.empty());
ASSERT_TRUE(variableRecords[0].optionalCast<DiscreteVariableRecord>());
DiscretePerturbationRecordVector perturbationRecords = variableRecords[0].cast<DiscreteVariableRecord>().discretePerturbationRecords(false);
EXPECT_EQ(5u,perturbationRecords.size());
ASSERT_TRUE(perturbationRecords.size() > 1);
EXPECT_TRUE(perturbationRecords[0].optionalCast<NullPerturbationRecord>());
EXPECT_TRUE(perturbationRecords[1].optionalCast<RubyPerturbationRecord>());
EXPECT_EQ(3u,analysisRecord.dataPointRecords().size());
EXPECT_TRUE(analysisRecord.completeDataPointRecords().empty());
}
analysis.clearDirtyFlag();
EXPECT_FALSE(analysis.isDirty());
EXPECT_FALSE(problem.isDirty());
BOOST_FOREACH(const Variable& variable, problem.variables()) {
EXPECT_FALSE(variable.isDirty());
ASSERT_TRUE(variable.optionalCast<DiscreteVariable>());
BOOST_FOREACH(const DiscretePerturbation& perturbation,variable.cast<DiscreteVariable>().perturbations(false)) {
EXPECT_FALSE(perturbation.isDirty());
}
}
BOOST_FOREACH(const Function& response, problem.responses()) {
EXPECT_FALSE(response.isDirty());
}
BOOST_FOREACH(const DataPoint& dataPoint, analysis.dataPoints()) {
EXPECT_FALSE(dataPoint.isDirty());
}
// make some more data points and make sure dirty flag bubbles up
values[0] = 3;
values[1] = 2;
values[2] = 2;
dataPoint = problem.createDataPoint(values);
ASSERT_TRUE(dataPoint);
test = analysis.addDataPoint(*dataPoint);
EXPECT_TRUE(test);
values[0] = 0;
values[1] = 4;
values[2] = 0;
dataPoint = problem.createDataPoint(values);
ASSERT_TRUE(dataPoint);
test = analysis.addDataPoint(*dataPoint);
EXPECT_TRUE(test);
EXPECT_EQ(5u,analysis.dataPoints().size());
EXPECT_TRUE(analysis.isDirty());
EXPECT_FALSE(problem.isDirty());
BOOST_FOREACH(const Variable& variable, problem.variables()) {
EXPECT_FALSE(variable.isDirty());
ASSERT_TRUE(variable.optionalCast<DiscreteVariable>());
BOOST_FOREACH(const DiscretePerturbation& perturbation,variable.cast<DiscreteVariable>().perturbations(false)) {
EXPECT_FALSE(perturbation.isDirty());
}
}
BOOST_FOREACH(const Function& response, problem.responses()) {
EXPECT_FALSE(response.isDirty());
}
int i = 0;
BOOST_FOREACH(const DataPoint& dataPoint, analysis.dataPoints()) {
if (i < 3) {
EXPECT_FALSE(dataPoint.isDirty());
}
else {
EXPECT_TRUE(dataPoint.isDirty());
}
++i;
}
// save to database, make sure changes made it in
{
bool transactionStarted = database.startTransaction();
EXPECT_TRUE(transactionStarted);
database.unloadUnusedCleanRecords(); // ETH@20130201 - Having to call this is awkward.
// Where is the best place?
AnalysisRecord analysisRecord(analysis,database);
database.save();
test = database.commitTransaction();
EXPECT_TRUE(test);
EXPECT_EQ(analysis.uuid(),analysisRecord.handle());
EXPECT_EQ(analysis.versionUUID(),analysisRecord.uuidLast());
ASSERT_NO_THROW(analysisRecord.problemRecord());
ProblemRecord problemRecord = analysisRecord.problemRecord();
EXPECT_EQ(problemRecord.handle(),analysisRecord.problemRecord().handle());
EXPECT_EQ(problemRecord.uuidLast(),analysisRecord.problemRecord().uuidLast());
InputVariableRecordVector variableRecords = problemRecord.inputVariableRecords();
EXPECT_EQ(3u,variableRecords.size());
ASSERT_FALSE(variableRecords.empty());
ASSERT_TRUE(variableRecords[0].optionalCast<DiscreteVariableRecord>());
DiscretePerturbationRecordVector perturbationRecords = variableRecords[0].cast<DiscreteVariableRecord>().discretePerturbationRecords(false);
EXPECT_EQ(5u,perturbationRecords.size());
ASSERT_TRUE(perturbationRecords.size() > 1);
EXPECT_TRUE(perturbationRecords[0].optionalCast<NullPerturbationRecord>());
EXPECT_TRUE(perturbationRecords[1].optionalCast<RubyPerturbationRecord>());
EXPECT_EQ(5u,analysisRecord.dataPointRecords().size());
EXPECT_TRUE(analysisRecord.completeDataPointRecords().empty());
}
analysis.clearDirtyFlag();
EXPECT_FALSE(analysis.isDirty());
EXPECT_FALSE(problem.isDirty());
BOOST_FOREACH(const Variable& variable, problem.variables()) {
EXPECT_FALSE(variable.isDirty());
ASSERT_TRUE(variable.optionalCast<DiscreteVariable>());
BOOST_FOREACH(const DiscretePerturbation& perturbation,variable.cast<DiscreteVariable>().perturbations(false)) {
EXPECT_FALSE(perturbation.isDirty());
}
}
BOOST_FOREACH(const Function& response, problem.responses()) {
EXPECT_FALSE(response.isDirty());
}
BOOST_FOREACH(const DataPoint& dataPoint, analysis.dataPoints()) {
EXPECT_FALSE(dataPoint.isDirty());
}
// remove a data point and save database. make sure point actually disappears
DataPoint toRemove = analysis.dataPoints()[3];
test = analysis.removeDataPoint(toRemove);
EXPECT_TRUE(test);
EXPECT_TRUE(analysis.isDirty());
{
bool transactionStarted = database.startTransaction();
EXPECT_TRUE(transactionStarted);
database.unloadUnusedCleanRecords(); // ETH@20130201 - Having to call this is awkward.
// Where is the best place?
AnalysisRecord analysisRecord(analysis,database);
database.save();
test = database.commitTransaction();
EXPECT_TRUE(test);
EXPECT_EQ(analysis.uuid(),analysisRecord.handle());
EXPECT_EQ(analysis.versionUUID(),analysisRecord.uuidLast());
ASSERT_NO_THROW(analysisRecord.problemRecord());
ProblemRecord problemRecord = analysisRecord.problemRecord();
EXPECT_EQ(problemRecord.handle(),analysisRecord.problemRecord().handle());
EXPECT_EQ(problemRecord.uuidLast(),analysisRecord.problemRecord().uuidLast());
InputVariableRecordVector variableRecords = problemRecord.inputVariableRecords();
EXPECT_EQ(3u,variableRecords.size());
ASSERT_FALSE(variableRecords.empty());
ASSERT_TRUE(variableRecords[0].optionalCast<DiscreteVariableRecord>());
DiscretePerturbationRecordVector perturbationRecords = variableRecords[0].cast<DiscreteVariableRecord>().discretePerturbationRecords(false);
EXPECT_EQ(5u,perturbationRecords.size());
ASSERT_TRUE(perturbationRecords.size() > 1);
EXPECT_TRUE(perturbationRecords[0].optionalCast<NullPerturbationRecord>());
EXPECT_TRUE(perturbationRecords[1].optionalCast<RubyPerturbationRecord>());
EXPECT_EQ(4u,analysisRecord.dataPointRecords().size());
OptionalDataPointRecord searchResult = database.getObjectRecordByHandle<DataPointRecord>(toRemove.uuid());
EXPECT_FALSE(searchResult);
EXPECT_EQ(4u,DataPointRecord::getDataPointRecords(database).size());
}
EXPECT_EQ(4u,analysis.dataPoints().size());
test = analysis.clearDirtyFlag();
EXPECT_TRUE(test);
}
EndCriteria::Type FireflyAlgorithm::minimize(Problem &P, const EndCriteria &endCriteria) {
QL_REQUIRE(!P.constraint().empty(), "Firefly Algorithm is a constrained optimizer");
EndCriteria::Type ecType = EndCriteria::None;
P.reset();
Size iteration = 0;
Size iterationStat = 0;
Size maxIteration = endCriteria.maxIterations();
Size maxIStationary = endCriteria.maxStationaryStateIterations();
startState(P, endCriteria);
bool isFA = Mfa_ > 0 ? true : false;
//Variables for DE
Array z(N_, 0.0);
Size indexBest, indexR1, indexR2;
//Set best value & position
Real bestValue = values_[0].first;
Size bestPosition = 0;
for (Size i = 1; i < M_; i++) {
if (values_[i].first < bestValue) {
bestPosition = i;
bestValue = values_[i].first;
}
}
Array bestX = x_[bestPosition];
//Run optimization
do {
iteration++;
iterationStat++;
//Check if stopping criteria is met
if (iteration > maxIteration || iterationStat > maxIStationary)
break;
//Divide into two subpopulations
//First sort values
std::sort(values_.begin(), values_.end());
//Differential evolution
if(Mfa_ < M_){
Size indexBest = values_[0].second;
Array& xBest = x_[indexBest];
for (Size i = Mfa_; i < M_; i++) {
if (!isFA) {
//Pure DE requires random index
indexBest = drawIndex_();
xBest = x_[indexBest];
}
do {
indexR1 = drawIndex_();
} while(indexR1 == indexBest);
do {
indexR2 = drawIndex_();
} while(indexR2 == indexBest || indexR2 == indexR1);
Size index = values_[i].second;
Array& x = x_[index];
Array& xR1 = x_[indexR1];
Array& xR2 = x_[indexR2];
for (Size j = 0; j < N_; j++) {
if (rng_.nextReal() <= crossover_) {
//Change x[j] according to crossover
z[j] = xBest[j] + mutation_*(xR1[j] - xR2[j]);
} else {
z[j] = x[j];
}
}
Real val = P.value(z);
if (val < values_[index].first) {
//Accept new point
x = z;
values_[index].first = val;
//mark best
if (val < bestValue) {
bestValue = val;
bestX = x;
iterationStat = 0;
}
}
}
}
//Firefly algorithm
if(isFA){
//According to the intensity, determine best global position
intensity_->findBrightest();
//Prepare random walk
randomWalk_->walk();
//Loop over particles
for (Size i = 0; i < Mfa_; i++) {
Size index = values_[i].second;
Array& x = x_[index];
Array& xI = xI_[index];
Array& xRW = xRW_[index];
//Loop over dimensions
for (Size j = 0; j < N_; j++) {
//Update position
x[j] += xI[j] + xRW[j];
//Enforce bounds on positions
if (x[j] < lX_[j]) {
x[j] = lX_[j];
}
else if (x[j] > uX_[j]) {
x[j] = uX_[j];
}
}
//Evaluate x & mark best
values_[index].first = P.value(x);
if (values_[index].first < bestValue) {
bestValue = values_[index].first;
bestX = x;
iterationStat = 0;
}
}
}
} while (true);
if (iteration > maxIteration)
ecType = EndCriteria::MaxIterations;
else
ecType = EndCriteria::StationaryPoint;
//Set result to best point
P.setCurrentValue(bestX);
P.setFunctionValue(bestValue);
return ecType;
}
Result solve(const Problem& iProblem) const {
// set up initial (empty) result
Result result;
result.mSuccess = false;
result.mNumIterations = 0;
// ensure that there are enough data points to proceed
const int sampleSize = iProblem.getSampleSize();
const int n = iProblem.getNumDataPoints();
if (n < sampleSize) {
return result;
}
const double epsilon = 1e-10;
// best results are currently invalid
double bestScore = -1;
bool success = false;
// start number of iterations as infinite, then reduce as we go
double numIterationsNeeded = 1e10;
int iterationCount = 0;
int skippedSampleCount = 0;
// for random sample index generation
std::vector<int> allIndices(n);
// iterate until adaptive number of iterations are exceeded
while (iterationCount < numIterationsNeeded) {
// determine random sample indices
for (int i = 0; i < n; ++i) {
allIndices[i] = i;
}
for (int i = 0; i < sampleSize; ++i) {
int randIndex = std::rand() % n;
std::swap(allIndices[i], allIndices[randIndex]);
}
std::vector<int> sampleIndices(allIndices.begin(),
allIndices.begin() + sampleSize);
// compute solution on minimal set
typename Problem::Solution solution = iProblem.estimate(sampleIndices);
// compute errors over all data points
std::vector<double> errors2 = iProblem.computeSquaredErrors(solution);
// check whether this is a valid sample
// TODO: this should be done via a method in Problem class, but would
// require changing all existing usages to include that method
if (errors2.size() == 0) {
++skippedSampleCount;
if (skippedSampleCount >= mMaximumIterations) break;
continue;
}
skippedSampleCount = 0;
// compute error threshold to be applied to each term
double thresh = mMaximumError;
if (thresh < 0) {
std::sort(errors2.begin(), errors2.end());
double median = (n % 2 == 0) ?
(0.5*(errors2[n/2]+errors2[n/2+1])) : errors2[n/2];
thresh = 1.4826*std::sqrt(median)*4.6851;
}
thresh *= thresh;
// determine inliers
std::vector<int> inliers;
inliers.reserve(n);
for (int i = 0; i < n; ++i) {
if (errors2[i] <= thresh) {
inliers.push_back(i);
}
}
// if this is the best score, update solution and convergence criteria
double score = inliers.size();
if (score > bestScore) {
bestScore = score;
result.mInliers = inliers;
result.mSolution = solution;
success = true;
double inlierProbability = double(inliers.size()) / n;
double anyOutlierProbability = 1 - pow(inlierProbability,sampleSize);
anyOutlierProbability = std::min(anyOutlierProbability, 1-epsilon);
anyOutlierProbability = std::max(anyOutlierProbability, epsilon);
numIterationsNeeded =
log(1-mGoodSolutionProbability) / log(anyOutlierProbability);
}
// bump up iteration count and terminate if it exceeds hard max
++iterationCount;
if (iterationCount > mMaximumIterations) {
break;
}
}
// finish off result params
result.mSuccess = success;
result.mNumIterations = iterationCount;
// refine result using all inliers if specified
if (result.mSuccess && mRefineUsingInliers) {
result.mSolution = iProblem.estimate(result.mInliers);
}
// done
return result;
}
int main ( int argc, char* argv[] )
{
/* parse command line arguments */
std::string anArg;
std::string modelRoot;
#if defined(USE_MPI) || defined(USE_MPI_PARPROC) || defined(USE_MPI_REGSOIL) || \
defined(USE_MPI_GEMS) || defined(USE_MPI_KRC)
#ifdef OGS_FEM_IPQC
int nb_ddc=0; //number of cores for DDC related processes
#endif
#endif
for( int i = 1; i < argc; i++ )
{
anArg = std::string( argv[i] );
if( anArg == "--help" || anArg == "-h")
{
std::cout << "Usage: ogs [MODEL_ROOT] [OPTIONS]\n"
<< "Where OPTIONS are:\n"
<< " -h [--help] print this message and exit\n"
<< " -b [--build-info] print build info and exit\n"
<< " --output-directory DIR put output files into DIR\n"
<< " --version print ogs version and exit" << "\n";
continue;
}
if( anArg == "--build-info" || anArg == "-b" )
{
std::cout << "ogs version: " << BuildInfo::OGS_VERSION << "\n"
<< "ogs date: " << BuildInfo::OGS_DATE << "\n";
std::cout << "git commit info: " << BuildInfo::GIT_COMMIT_INFO << "\n";
std::cout << "build timestamp: " << BuildInfo::BUILD_TIMESTAMP << "\n";
continue;
}
if( anArg == "--version" )
{
std::cout << BuildInfo::OGS_VERSION << "\n";
continue;
}
if( anArg == "--model-root" || anArg == "-m" )
{
if (i+1 >= argc) {
std::cerr << "Error: Parameter " << anArg << " needs an additional argument" << std::endl;
std::exit(EXIT_FAILURE);
}
modelRoot = std::string( argv[++i] );
continue;
}
if (anArg == "--output-directory")
{
if (i+1 >= argc) {
std::cerr << "Error: Parameter " << anArg << " needs an additional argument" << std::endl;
std::exit(EXIT_FAILURE);
}
std::string path = argv[++i];
if (! path.empty()) defaultOutputPath = path;
continue;
}
#if defined(USE_MPI) || defined(USE_MPI_PARPROC) || defined(USE_MPI_REGSOIL) || \
defined(USE_MPI_GEMS) || defined(USE_MPI_KRC)
#ifdef OGS_FEM_IPQC
std::string decompositions;
if( anArg == "--domain-decomposition" || anArg == "-ddc" )
{
decompositions = std::string( argv[++i] );
nb_ddc = atoi(decompositions.c_str());
continue;
}
#endif
#endif
// anything left over must be the model root, unless already found
if ( modelRoot == "" )
modelRoot = std::string( argv[i] );
} // end of parse argc loop
if( argc > 1 && modelRoot == "" ) // non-interactive mode and no model given
exit(0); // e.g. just wanted the build info
std::string solver_pkg_name = BuildInfo::SOLVER_PACKAGE_NAME;
// No default linear solver package is in use.
if(solver_pkg_name.find("Default") == std::string::npos)
{
std::cout << "\nWarning: " << solver_pkg_name
<< " other than the OGS default one is in use." <<std::endl;
std::cout << " The solver setting may need to be adjusted for the solution accuracy!" << std::endl;
}
char* dateiname(NULL);
#ifdef SUPERCOMPUTER
// *********************************************************************
// buffered output ... important for performance on cray
// (unbuffered output is limited to 10 bytes per second)
// [email protected] 11.10.2007
char buf[1024 * 1024];
int bsize;
bsize = 1024 * 1024; // question: what happens if buffer is full?
// according to documentation the buffer is flushed when full.
// If we have a lot of output, increasing buffer is usefull.
if(bsize > 0)
// bufstd = malloc(bsize);
setvbuf(stdout, buf, _IOFBF, bsize);
//**********************************************************************
#endif
/*---------- MPI Initialization ----------------------------------*/
#if defined(USE_MPI) || defined(USE_MPI_PARPROC) || defined(USE_MPI_REGSOIL) || \
defined(USE_MPI_GEMS) || defined(USE_MPI_KRC)
printf("Before MPI_Init\n");
#if defined(USE_MPI_GEMS)
int prov;
MPI_Init_thread(&argc,&argv,MPI_THREAD_FUNNELED, &prov);
#else
MPI_Init(&argc,&argv);
#endif
MPI_Barrier (MPI_COMM_WORLD); // 12.09.2007 WW
elapsed_time_mpi = -MPI_Wtime(); // 12.09.2007 WW
#ifdef OGS_FEM_IPQC
bool splitcomm_flag;
int np;
MPI_Comm_size(MPI_COMM_WORLD, &np);
splitcomm_flag = SplitMPI_Communicator::CreateCommunicator(MPI_COMM_WORLD, np, nb_ddc);
time_ele_paral = 0.0;
#else
comm_DDC= MPI_COMM_WORLD;
#endif
#endif
/*---------- MPI Initialization ----------------------------------*/
#ifdef USE_PETSC
int rank, r_size;
PetscLogDouble v1,v2;
char help[] = "OGS with PETSc \n";
//PetscInitialize(argc, argv, help);
PetscInitialize(&argc,&argv,(char *)0,help);
//kg44 quick fix to compile PETSC with version PETSCV3.4
#ifdef USEPETSC34
PetscTime(&v1);
#else
PetscGetTime(&v1);
#endif
MPI_Comm_rank(PETSC_COMM_WORLD, &rank);
MPI_Comm_size(PETSC_COMM_WORLD, &r_size);
PetscSynchronizedPrintf(PETSC_COMM_WORLD, "===\nUse PETSc solver");
PetscSynchronizedPrintf(PETSC_COMM_WORLD, "Number of CPUs: %d, rank: %d\n", r_size, rank);
#endif
/*---------- LIS solver -----------------------------------------*/
#ifdef LIS
//Initialization of the lis solver.
lis_initialize(&argc, &argv);
#endif
/*========================================================================*/
/* Kommunikation mit Betriebssystem */
/* Timer fuer Gesamtzeit starten */
#ifdef TESTTIME
TStartTimer(0);
#endif
/* Intro ausgeben */
#if defined(USE_MPI) //WW
if(myrank == 0)
#endif
#ifdef USE_PETSC
if(rank == 0 )
#endif
DisplayStartMsg();
/* Speicherverwaltung initialisieren */
if (!InitMemoryTest())
{
DisplayErrorMsg("Fehler: Speicherprotokoll kann nicht erstellt werden!");
DisplayErrorMsg(" Programm vorzeitig beendet!");
return 1; // LB changed from 0 to 1 because 0 is indicating success
}
if( argc == 1 ) // interactive mode
dateiname = ReadString();
else // non-interactive mode
{
if ( argc == 2 ) // a model root was supplied
{
dateiname = (char*) Malloc((int)strlen(argv[1]) + 1);
dateiname = strcpy(dateiname,argv[1]);
}
else // several args supplied
if( modelRoot != "")
{
dateiname = (char*) Malloc( (int) modelRoot.size() + 1 );
dateiname = strcpy( dateiname, modelRoot.c_str() );
}
DisplayMsgLn(dateiname);
}
//WW DisplayMsgLn("");
//WW DisplayMsgLn("");
// ----------23.02.2009. WW-----------------
// LB Check if file exists
std::string tmpFilename = dateiname;
tmpFilename.append(".pcs");
if(!IsFileExisting(tmpFilename))
{
std::cout << " Error: Cannot find file " << dateiname << "\n";
return 1;
}
// If no option is given, output files are placed in the same directory as the input files
if (defaultOutputPath.empty()) defaultOutputPath = pathDirname(std::string(dateiname));
FileName = dateiname;
size_t indexChWin, indexChLinux;
indexChWin = indexChLinux = 0;
indexChWin = FileName.find_last_of('\\');
indexChLinux = FileName.find_last_of('/');
//
if(indexChWin != std::string::npos)
FilePath = FileName.substr(0,indexChWin) + "\\";
else if(indexChLinux != std::string::npos)
FilePath = FileName.substr(0,indexChLinux) + "/";
// ---------------------------WW
Problem* aproblem = new Problem(dateiname);
#ifdef USE_PETSC
aproblem->setRankandSize(rank, r_size);
#endif
#if defined(USE_MPI) || defined(USE_MPI_PARPROC) || defined(USE_MPI_REGSOIL) || defined(USE_MPI_GEMS) || defined(USE_MPI_KRC)
aproblem->setRankandSize(myrank, mysize);
if (myrank != MPI_UNDEFINED)
{
#endif
aproblem->Euler_TimeDiscretize();
delete aproblem;
aproblem = NULL;
#if defined(USE_MPI) || defined(USE_MPI_PARPROC) || defined(USE_MPI_REGSOIL) || defined(USE_MPI_GEMS) || defined(USE_MPI_KRC)
}
#ifdef OGS_FEM_IPQC
//sending killing signals to ranks of group_IPQC, only when the group exists
if (splitcomm_flag == true) {
int signal = -1, rank_IPQC, mysize_IPQC = np - nb_ddc;
for (int i=0; i< mysize_IPQC; i++) {
rank_IPQC = mysize + i;
MPI_Send(&signal, 1, MPI_INT, rank_IPQC, 0, MPI_COMM_WORLD);
}
}
#endif
#endif
if(ClockTimeVec.size()>0)
ClockTimeVec[0]->PrintTimes(); //CB time
DestroyClockTime();
#ifdef TESTTIME
#if defined(USE_MPI)
if(myrank == 0)
#endif
#if defined(USE_PETSC)
if(rank == 0)
#endif
std::cout << "Simulation time: " << TGetTimer(0) << "s" << "\n";
#endif
/* Abspann ausgeben */
/*--------- MPI Finalize ------------------*/
#if defined(USE_MPI) || defined(USE_MPI_PARPROC) || defined(USE_MPI_REGSOIL) || defined(USE_MPI_KRC)
elapsed_time_mpi += MPI_Wtime(); // 12.09.2007 WW
std::cout << "\n *** Total CPU time of parallel modeling: " << elapsed_time_mpi <<
"\n"; //WW
// Count CPU time of post time loop WW
MPI_Finalize();
#endif
/*--------- MPI Finalize ------------------*/
/*--------- LIS Finalize ------------------*/
#ifdef LIS
lis_finalize();
#endif
/*--------- LIS Finalize ------------------*/
free(dateiname);
#ifdef USE_PETSC
//kg44 quick fix to compile PETSC with version PETSCV3.4
#ifdef USEPETSC34
PetscTime(&v2);
#else
PetscGetTime(&v2);
#endif
PetscPrintf(PETSC_COMM_WORLD,"\t\n>>Total elapsed time by using PETSC:%f s\n",v2-v1);
PetscFinalize();
#endif
return 0;
}
void TDefense::timerEvent(QTimerEvent *event)
{
//if (isLaunched)
//{
std::cout << "\nWeek: " << week;
launchModerator++;
int round = week - 1;
if (numEnemiesLaunched <= 10 + (round * 10))
{
if (launchModerator % 50 - (round * 5) == 0)
{
Problem *p = new Problem(gpWidth / 2, 0, problem);
myProblems.push_back(p);
numEnemiesLaunched++;
}
if (week > 1)
{
Tower *r = new Tower(500, 300, stapler);
myTowers.push_back(r);
if (launchModerator % 80 - (round * 10) == 0)
{
Assignment *a = new Assignment(gpWidth / 2, 0, assignment1);
myAssignments.push_back(a);
numEnemiesLaunched++;
} // end 'launchModerator' if
} // end 'week' if
if (week % 3 == 0)
{
if (launchModerator % 150 - (round * 15) == 0)
{
Paper *r = new Paper(gpWidth / 2, 0, paper1);
myPapers.push_back(r);
numEnemiesLaunched++;
} // end 'launchModerator' if
} // 'week' if
if (week == 5 || week == 9)
{
if (launchModerator % 300 - (round * 20) == 0)
{
Midterm *m = new Midterm(gpWidth / 2, 0, midterm5);
myMidterms.push_back(m);
numEnemiesLaunched++;
} // end 'launchModerator' if
} // 'week' if
} // end 'numEnemiesLaunched' if
enemyAction();
if (myProblems.size() > 0) {
Problem *curProblem = myProblems.back();
std::cout << "myProblems size: " << myProblems.size();
std::cout << "\nproblem y-pos: " << curProblem->getY();
}
towerAction();
if (studentShotFired)
studentAction();
if (extensions <= 0)
gameOver();
if ((numEnemiesShot + numEnemiesDropped) >= 10 + (round * 10))
{
clearEnemies();
numEnemiesDropped = 0;
numEnemiesLaunched = 1;
numEnemiesShot = 0;
isLaunched = false;
IQ += 10;
week++;
}
if (week == 10)
semester++;
if (semester == 2)
{
semester = 1;
year++;
}
repaint();
//} // end 'isLaunch' if
} // end method
void TDefense::paintEvent(QPaintEvent *event)
{
QPainter painter(this);
QColor* color = new QColor(255, 255, 255);
painter.fillRect(*windowRect, *color);
painter.drawRect(*windowRect);
if (isOver) {
QFont font("Courier", 15, QFont::DemiBold);
QFontMetrics fm(font);
int textWidth = fm.width("GAME OVER");
painter.setFont(font);
int h = height();
int w = width();
painter.translate(QPoint(w/2, h/2));
painter.drawText(-textWidth/2, 0, "GAME OVER");
} else if(isWon) {
QFont font("Courier", 15, QFont::DemiBold);
QFontMetrics fm(font);
int textWidth = fm.width("VICTORY");
painter.setFont(font);
int h = height();
int w = width();
painter.translate(QPoint(w/2, h/2));
painter.drawText(-textWidth/2, 0, "VICTORY");
} else if (isPaused) {
QFont font("Courier", 15, QFont::DemiBold);
QFontMetrics fm(font);
int textWidth = fm.width("PAUSED");
painter.setFont(font);
int h = height();
int w = width();
painter.translate(QPoint(w/2, h/2));
painter.drawText(-textWidth/2, 0, "PAUSED");
}
else
{
QFont font("Brush Script MT", 15);
painter.setFont(font);
painter.translate(QPoint(30, 30));
painter.drawText(0, 0, "WWeek: " + week);
painter.translate(QPoint(13, 15));
painter.drawText(0, 0, "SSemester: " + semester);
painter.translate(QPoint(30, 15));
painter.drawText(0, 0, "YYear: " + year);
painter.translate(QPoint(-50, 300));
painter.drawText(0, 0, "IQ: ");
painter.translate(QPoint(10, 17));
painter.drawText(0, 0, "Funds: " + funds);
painter.translate(QPoint(10, 17));
painter.drawText(0, 0, "EExtensionExtensions: " + extensions);
painter.drawImage(myStudent->getRect()->center(), myStudent->getImage());
/*for (int i = 0; i < myTowers.size(); i++)
{
Tower *curTower = myTowers[i];
painter.drawImage(*curTower->getRect(), curTower->getImage());
}*/
//if (isLaunched)
//{
for (int i = 0; i < myProblems.size(); i++)
{
Problem *curProblem = myProblems[i];
painter.drawImage(*curProblem->getRect(), curProblem->getImage());
} // end for
/*
for (int i = 0; i < myAssignments.size(); i++)
{
Assignment *curAssignment = myAssignments[i];
painter.drawImage(*curAssignment->getRect(), curAssignment->getImage());
} // end for
for (int i = 0; i < myPapers.size(); i++)
{
Paper *curPaper = myPapers[i];
painter.drawImage(*curPaper->getRect(), curPaper->getImage());
} // end for
for (int i = 0; i < myMidterms.size(); i++)
{
Midterm *curMidterm = myMidterms[i];
painter.drawImage(*curMidterm->getRect(), curMidterm->getImage());
} // end for
*/
if (studentShotFired)
{
for (int i = 0; i < myStudentBullets.size(); i++)
{
Bullet *curBullet = myStudentBullets[i];
painter.drawImage(*curBullet->getRect(), curBullet->getImage());
} // end for
} // end 'shotFired' if
for (int i = 0; i < myTowerBullets.size(); i++)
{
Bullet *curBullet = myTowerBullets[i];
painter.drawImage(*curBullet->getRect(), curBullet->getImage());
} // end for
//} // end 'isLaunch' if
} // end 'isPaused' else
}
// --------------------------------------------------------------------------------------------------------------------
// --------------------------------------------------------------------------------------------------------------------
// --------------------------------------------------------------------------------------------------------------------
ProblemFactory::ProblemFactory(bool _skipInit)
{
Problem* newProb = nullptr;
// Null
newProb = new NullProblem();
if (_skipInit || newProb->Init()) {
if (!_skipInit) {
newProb->Shutdown();
}
mProblems.push_back(newProb);
mSolutions[mProblems.back()->GetName()].push_back(new NullSolution());
} else {
newProb->Shutdown();
SafeDelete(newProb);
console::error("Unable to create the Null Problem--exiting.");
}
// DynamicStreaming
newProb = new DynamicStreamingProblem();
if (_skipInit || newProb->Init()) {
if (!_skipInit) {
newProb->Shutdown();
}
mProblems.push_back(newProb);
mSolutions[mProblems.back()->GetName()].push_back(new DynamicStreamingGLBufferSubData());
mSolutions[mProblems.back()->GetName()].push_back(new DynamicStreamingGLMapUnsynchronized());
mSolutions[mProblems.back()->GetName()].push_back(new DynamicStreamingGLMapPersistent());
#if WITH_D3D11
mSolutions[mProblems.back()->GetName()].push_back(new DynamicStreamingD3D11MapNoOverwrite());
mSolutions[mProblems.back()->GetName()].push_back(new DynamicStreamingD3D11UpdateSubresource());
#endif
} else {
newProb->Shutdown();
SafeDelete(newProb);
}
// UntexturedObjects
newProb = new UntexturedObjectsProblem();
if (_skipInit || newProb->Init()) {
if (!_skipInit) {
newProb->Shutdown();
}
mProblems.push_back(newProb);
mSolutions[mProblems.back()->GetName()].push_back(new UntexturedObjectsGLUniform());
mSolutions[mProblems.back()->GetName()].push_back(new UntexturedObjectsGLDrawLoop());
mSolutions[mProblems.back()->GetName()].push_back(new UntexturedObjectsGLMultiDraw(true));
mSolutions[mProblems.back()->GetName()].push_back(new UntexturedObjectsGLMultiDraw(false));
mSolutions[mProblems.back()->GetName()].push_back(new UntexturedObjectsGLMultiDrawBuffer(true));
mSolutions[mProblems.back()->GetName()].push_back(new UntexturedObjectsGLMultiDrawBuffer(false));
mSolutions[mProblems.back()->GetName()].push_back(new UntexturedObjectsGLBindless());
mSolutions[mProblems.back()->GetName()].push_back(new UntexturedObjectsGLBindlessIndirect());
mSolutions[mProblems.back()->GetName()].push_back(new UntexturedObjectsGLBufferRange());
mSolutions[mProblems.back()->GetName()].push_back(new UntexturedObjectsGLBufferStorage(true));
mSolutions[mProblems.back()->GetName()].push_back(new UntexturedObjectsGLBufferStorage(false));
mSolutions[mProblems.back()->GetName()].push_back(new UntexturedObjectsGLDynamicBuffer());
mSolutions[mProblems.back()->GetName()].push_back(new UntexturedObjectsGLMapUnsynchronized());
mSolutions[mProblems.back()->GetName()].push_back(new UntexturedObjectsGLMapPersistent());
mSolutions[mProblems.back()->GetName()].push_back(new UntexturedObjectsGLTexCoord());
#if WITH_D3D11
mSolutions[mProblems.back()->GetName()].push_back(new UntexturedObjectsD3D11Naive());
#endif
} else {
newProb->Shutdown();
SafeDelete(newProb);
}
// Textured Quads
newProb = new TexturedQuadsProblem();
if (_skipInit || newProb->Init()) {
if (!_skipInit) {
newProb->Shutdown();
}
mProblems.push_back(newProb);
mSolutions[mProblems.back()->GetName()].push_back(new TexturedQuadsGLBindless());
mSolutions[mProblems.back()->GetName()].push_back(new TexturedQuadsGLBindlessMultiDraw());
mSolutions[mProblems.back()->GetName()].push_back(new TexturedQuadsGLNaive());
mSolutions[mProblems.back()->GetName()].push_back(new TexturedQuadsGLNaiveUniform());
mSolutions[mProblems.back()->GetName()].push_back(new TexturedQuadsGLNoTex());
mSolutions[mProblems.back()->GetName()].push_back(new TexturedQuadsGLNoTexUniform());
mSolutions[mProblems.back()->GetName()].push_back(new TexturedQuadsGLSparseBindlessTextureArray());
mSolutions[mProblems.back()->GetName()].push_back(new TexturedQuadsGLSparseBindlessTextureArrayMultiDraw(true));
mSolutions[mProblems.back()->GetName()].push_back(new TexturedQuadsGLSparseBindlessTextureArrayMultiDraw(false));
mSolutions[mProblems.back()->GetName()].push_back(new TexturedQuadsGLTextureArray());
mSolutions[mProblems.back()->GetName()].push_back(new TexturedQuadsGLTextureArrayUniform());
mSolutions[mProblems.back()->GetName()].push_back(new TexturedQuadsGLTextureArrayMultiDraw(true));
mSolutions[mProblems.back()->GetName()].push_back(new TexturedQuadsGLTextureArrayMultiDraw(false));
mSolutions[mProblems.back()->GetName()].push_back(new TexturedQuadsGLTextureArrayMultiDrawBuffer(true));
mSolutions[mProblems.back()->GetName()].push_back(new TexturedQuadsGLTextureArrayMultiDrawBuffer(false));
#if WITH_D3D11
mSolutions[mProblems.back()->GetName()].push_back(new TexturedQuadsD3D11Naive());
#endif
} else {
newProb->Shutdown();
SafeDelete(newProb);
}
}
Real GoldsteinLineSearch::operator()(Problem& P,
EndCriteria::Type& ecType,
const EndCriteria& endCriteria,
const Real t_ini)
{
Constraint& constraint = P.constraint();
succeed_=true;
bool maxIter = false;
Real qtold, t = t_ini;
Size loopNumber = 0;
Real q0 = P.functionValue();
Real qp0 = P.gradientNormValue();
Real tl = 0.0;
Real tr = 0.0;
qt_ = q0;
qpt_ = (gradient_.empty()) ? qp0 : -DotProduct(gradient_,searchDirection_);
// Initialize gradient
gradient_ = Array(P.currentValue().size());
// Compute new point
xtd_ = P.currentValue();
t = update(xtd_, searchDirection_, t, constraint);
// Compute function value at the new point
qt_ = P.value (xtd_);
while ((qt_ - q0) < -beta_*t*qpt_ || (qt_ - q0) > -alpha_*t*qpt_) {
if ((qt_ - q0) > -alpha_*t*qpt_)
tr = t;
else
tl = t;
++loopNumber;
// calculate the new step
if (close_enough(tr, 0.0))
t *= extrapolation_;
else
t = (tl + tr) / 2.0;
// Store old value of the function
qtold = qt_;
// New point value
xtd_ = P.currentValue();
t = update(xtd_, searchDirection_, t, constraint);
// Compute function value at the new point
qt_ = P.value (xtd_);
P.gradient (gradient_, xtd_);
// and it squared norm
maxIter = endCriteria.checkMaxIterations(loopNumber, ecType);
if (maxIter)
break;
}
if (maxIter)
succeed_ = false;
// Compute new gradient
P.gradient(gradient_, xtd_);
// and it squared norm
qpt_ = DotProduct(gradient_, gradient_);
// Return new step value
return t;
}
/**
* Class implementing the NSGA-II algorithm.
* This implementation of NSGA-II makes use of a QualityIndicator object
* to obtained the convergence speed of the algorithm. This version is used
* in the paper:
* A.J. Nebro, J.J. Durillo, C.A. Coello Coello, F. Luna, E. Alba
* "A Study of Convergence Speed in Multi-Objective Metaheuristics."
* To be presented in: PPSN'08. Dortmund. September 2008.
*
* Besides the classic NSGA-II, a steady-state version (ssNSGAII) is also
* included (See: J.J. Durillo, A.J. Nebro, F. Luna and E. Alba
* "On the Effect of the Steady-State Selection Scheme in
* Multi-Objective Genetic Algorithms"
* 5th International Conference, EMO 2009, pp: 183-197.
* April 2009)
*/
int main(int argc, char ** argv) {
clock_t t_ini, t_fin;
Problem * problem ; // The problem to solve
Algorithm * algorithm ; // The algorithm to use
HUXCrossover * crossover ; // Crossover operator
BitFlipMutation * mutation ; // Mutation operator
BinaryTournament * selection ; // Selection operator
if (argc>=2) {
problem = ProblemFactory::getProblem(argc, argv);
cout << "Selected problem: " << problem->getName() << endl;
} else {
cout << "No problem selected." << endl;
cout << "Default problem will be used: Fonseca" << endl;
problem = ProblemFactory::getProblem(const_cast<char *>("ZDT5"));
}
// QualityIndicator * indicators ; // Object to get quality indicators
// indicators = NULL ;
algorithm = new MOCHC(problem);
// Algorithm parameters
int populationSize = 100;
int maxEvaluations = 25000;
double initialConvergenceCount = 0.25;
double preservedPopulation = 0.05;
int convergenceValue = 3;
algorithm->setInputParameter("populationSize",&populationSize);
algorithm->setInputParameter("maxEvaluations",&maxEvaluations);
algorithm->setInputParameter("initialConvergenceCount",&initialConvergenceCount);
algorithm->setInputParameter("preservedPopulation",&preservedPopulation);
algorithm->setInputParameter("convergenceValue",&convergenceValue);
// Mutation and Crossover for Real codification
map<string, void *> parameters;
double crossoverProbability = 1.0;
double crossoverDistributionIndex = 20.0;
parameters["probability"] = &crossoverProbability;
crossover = new HUXCrossover(parameters);
parameters.clear();
double mutationProbability = 0.35;
parameters["probability"] = &mutationProbability;
mutation = new BitFlipMutation(parameters);
// Selection Operator
parameters.clear();
selection = new BinaryTournament(parameters);
// Add the operators to the algorithm
algorithm->addOperator("crossover",crossover);
algorithm->addOperator("mutation",mutation);
algorithm->addOperator("parentSelection",selection);
// Add the indicator object to the algorithm
// algorithm->setInputParameter("indicators", indicators) ;
// Execute the Algorithm
t_ini = clock();
SolutionSet * population = algorithm->execute();
t_fin = clock();
double secs = (double) (t_fin - t_ini);
secs = secs / CLOCKS_PER_SEC;
// Result messages
cout << "Total execution time: " << secs << "s" << endl;
cout << "Variables values have been written to file VAR" << endl;
population->printVariablesToFile("VAR");
cout << "Objectives values have been written to file FUN" << endl;
population->printObjectivesToFile("FUN");
delete selection;
delete mutation;
delete crossover;
delete population;
delete algorithm;
} // main
void lidarBoostEngine::build_superresolution(short coeff)
{
std::cout<< "Num of clouds : " << Y.size() << std::endl;
// std::cout << Y[0] << std::endl;
beta = coeff;
std::vector < MatrixXd > optflow = lk_optical_flow( Y[2], Y[4], 10 );
MatrixXd D( beta*n, beta*m ); //, X( beta*n, beta*m );
// SparseMatrix<double> W( beta*n, beta*m ), T( beta*n, beta*m );
D = apply_optical_flow(Y[2], optflow);
T = check_unreliable_samples(intensityMap[2], 0.0001);
MatrixXd up_D = nearest_neigh_upsampling(D);
//// Display and Debug
cv::Mat M(n, m, CV_32FC1);
// MatrixXd diff1(n, m);
// diff1 = MatrixXd::Ones(n, m) - Y[0];
cv::eigen2cv(Y[2], M);
cv::Mat M1(n, m, CV_32FC1);
cv::eigen2cv(Y[4], M1);
// MatrixXd diff(beta*n, beta*m);
// diff = MatrixXd::Ones(beta*n, beta*m) - up_D;
cv::Mat M2(beta*n, beta*m, CV_32FC1);
cv::eigen2cv(up_D, M2);
cv::namedWindow("check", cv::WINDOW_AUTOSIZE );
cv::imshow("check", M);
cv::namedWindow("check1", cv::WINDOW_AUTOSIZE );
cv::imshow("check1", M1);
cv::namedWindow("check2", cv::WINDOW_AUTOSIZE );
cv::imshow("check2", M2);
//// Solve example equation with eigen
// Eigen::VectorXd x(2);
// x(0) = 10.0;
// x(1) = 25.0;
// std::cout << "x: " << x << std::endl;
// my_functor functor;
// Eigen::NumericalDiff<my_functor> numDiff(functor);
// Eigen::LevenbergMarquardt<Eigen::NumericalDiff<my_functor>,double> lm(numDiff);
// lm.parameters.maxfev = 2000;
// lm.parameters.xtol = 1.0e-10;
// std::cout << lm.parameters.maxfev << std::endl;
// int ret = lm.minimize(x);
// std::cout << lm.iter << std::endl;
// std::cout << ret << std::endl;
// std::cout << "x that minimizes the function: " << x << std::endl;
////// Try to solve lidarboost with Eigen
// my_functor functor;
// Eigen::NumericalDiff<my_functor> numDiff(functor);
// Eigen::LevenbergMarquardt<Eigen::NumericalDiff<my_functor>,double> lm(numDiff);
// lm.parameters.maxfev = 2000;
// lm.parameters.xtol = 1.0e-10;
// std::cout << lm.parameters.maxfev << std::endl;
// VectorXd val(2);
// for(int i = 0; i < X.rows(); i++)
// {
// for(int j = 0; j < X.cols(); j++)
// {
// val = X(i, j);
// int ret = lm.minimize(val);
// }
// }
// std::cout << lm.iter << std::endl;
// std::cout << ret << std::endl;
// std::cout << "x that minimizes the function: " << X << std::endl;
//// Solve example using ceres
// The variable to solve for with its initial value.
// double initial_x = 5.0;
// double x = initial_x;
MatrixXd X(beta*n, beta*m);// init_X(beta*n, beta*m);
// X = MatrixXd::Zero(beta*n,beta*m);
X = up_D;
// MatrixXd init_X( beta*n, beta*m );
// init_X = X;
// int M[2][2], M2[2][2];
// M[0][0] = 5;
// M[1][0] = 10;
// M[0][1] = 20;
// M[1][1] = 30;
// M2 = *M;
// Build the problem.
Problem problem;
// Set up the only cost function (also known as residual). This uses
// auto-differentiation to obtain the derivative (jacobian).
double val, w, t, d;
Solver::Options options;
options.linear_solver_type = ceres::DENSE_QR;
options.minimizer_progress_to_stdout = false;
Solver::Summary summary;
for(int i = 0; i < X.rows(); i++)
{
for(int j = 0; j < X.cols(); j++)
{
val = X(i, j);
w = W(i, j);
t = T(i, j);
d = up_D(i, j);
std::cout << "i = " << i << "; j = " << j << std::endl;
std::cout << "w = " << w << "; t = " << t << "; d = " << d << std::endl;
CostFunction* cost_function =
new AutoDiffCostFunction<CostFunctor, 1, 1>(new CostFunctor(w, t, d));
problem.AddResidualBlock(cost_function, NULL, &val);
// Run the solver
Solve(options, &problem, &summary);
X(i, j) = val;
}
}
std::cout << summary.BriefReport() << "\n";
// std::cout << "x : " << init_X
// << " -> " << X << "\n";
cv::Mat M3(beta*n, beta*m, CV_32FC1);
cv::eigen2cv(X, M3);
cv::namedWindow("check3", cv::WINDOW_AUTOSIZE );
cv::imshow("check3", M3);
}
int main(int argc, char ** argv) {
clock_t t_ini, t_fin;
Problem * problem ; // The problem to solve
Algorithm * algorithm ; // The algorithm to use
Operator * crossover ; // Crossover operator
Operator * selection ; // Selection operator
map<string, void *> parameters;
//TODO: QualityIndicator * indicators;
if (argc>=2) {
problem = ProblemFactory::getProblem(argc, argv);
cout << "Selected problem: " << problem->getName() << endl;
} else {
cout << "No problem selected." << endl;
cout << "Default problem will be used: Fonseca" << endl;
problem = ProblemFactory::getProblem(const_cast<char *>("Fonseca"));
}
algorithm = new GDE3(problem);
// Algorithm parameters
int populationSizeValue = 100;
int maxIterationsValue = 250;
algorithm->setInputParameter("populationSize",&populationSizeValue);
algorithm->setInputParameter("maxIterations",&maxIterationsValue);
// Crossover operator
double crParameter = 0.5;
double fParameter = 0.5;
parameters["CR"] = &crParameter;
parameters["F"] = &fParameter;
crossover = new DifferentialEvolutionCrossover(parameters);
// Selection operator
parameters.clear();
selection = new DifferentialEvolutionSelection(parameters) ;
// Add the operators to the algorithm
algorithm->addOperator("crossover",crossover);
algorithm->addOperator("selection",selection);
// Add the indicator object to the algorithm
//algorithm->setInputParameter("indicators", indicators) ;
// Execute the Algorithm
t_ini = clock();
SolutionSet * population = algorithm->execute();
t_fin = clock();
double secs = (double) (t_fin - t_ini);
secs = secs / CLOCKS_PER_SEC;
// Result messages
cout << "Total execution time: " << secs << "s" << endl;
cout << "Variables values have been written to file VAR" << endl;
population->printVariablesToFile("VAR");
cout << "Objectives values have been written to file FUN" << endl;
population->printObjectivesToFile("FUN");
delete selection;
delete crossover;
delete population;
delete algorithm;
} // main
int main(int argc, char ** argv) {
clock_t t_ini, t_fin;
Problem * problem; // The problem to solve
Algorithm * algorithm; // The algorithm to use
Operator * mutation; // "Turbulence" operator
map<string, void *> parameters; // Operator parameters
//TODO: QualityIndicator * indicators; // Object to get quality indicators
if (argc>=2) {
problem = ProblemFactory::getProblem(argc, argv);
cout << "Selected problem: " << problem->getName() << endl;
} else {
cout << "No problem selected." << endl;
cout << "Default problem will be used: Kursawe" << endl;
problem = ProblemFactory::getProblem(const_cast<char *>("Kursawe"));
}
algorithm = new SMPSO(problem);
// Algorithm parameters
int swarmSizeValue = 100;
int archiveSizeValue = 100;
int maxIterationsValue = 250;
algorithm->setInputParameter("swarmSize",&swarmSizeValue);
algorithm->setInputParameter("archiveSize",&archiveSizeValue);
algorithm->setInputParameter("maxIterations",&maxIterationsValue);
// Mutation operator
double probabilityParameter = 1.0/(problem->getNumberOfVariables());
double distributionIndexParameter = 20.0;
parameters["probability"] = &probabilityParameter;
parameters["distributionIndex"] = &distributionIndexParameter;
mutation = new PolynomialMutation(parameters);
// Add the operators to the algorithm
algorithm->addOperator("mutation",mutation);
// Add the indicator object to the algorithm
//algorithm->setInputParameter("indicators", indicators) ;
// Execute the Algorithm
t_ini = clock();
SolutionSet * population = algorithm->execute();
t_fin = clock();
double secs = (double) (t_fin - t_ini);
secs = secs / CLOCKS_PER_SEC;
// Result messages
cout << "Total execution time: " << secs << "s" << endl;
cout << "Variables values have been written to file VAR" << endl;
population->printVariablesToFile("VAR");
cout << "Objectives values have been written to file FUN" << endl;
population->printObjectivesToFile("FUN");
delete mutation;
delete population;
delete algorithm;
} // main
int main(int argc, char ** argv) {
clock_t t_ini, t_fin;
Problem * problem; // The problem to solve
Algorithm * algorithm; // The algorithm to use
Operator * crossover; // Crossover operator
Operator * mutation; // Mutation operator
//QualityIndicator * indicators ; // Object to get quality indicators
map<string, void *> parameters; // Operator parameters
//TODO: QualityIndicator * indicators; // Object to get quality indicators
if (argc>=2) {
problem = ProblemFactory::getProblem(argc, argv);
cout << "Selected problem: " << problem->getName() << endl;
} else {
cout << "No problem selected." << endl;
cout << "Default problem will be used: Kursawe" << endl;
problem = ProblemFactory::getProblem(const_cast<char *>("Kursawe"));
}
algorithm = new MOEAD(problem);
// Algorithm parameters
int populationSizeValue = 300;
int maxEvaluationsValue = 150000;
algorithm->setInputParameter("populationSize",&populationSizeValue);
algorithm->setInputParameter("maxEvaluations",&maxEvaluationsValue);
// Directory with the files containing the weight vectors used in
// Q. Zhang, W. Liu, and H Li, The Performance of a New Version of MOEA/D
// on CEC09 Unconstrained MOP Test Instances Working Report CES-491, School
// of CS & EE, University of Essex, 02/2009.
// http://dces.essex.ac.uk/staff/qzhang/MOEAcompetition/CEC09final/code/ZhangMOEADcode/moead0305.rar
string dataDirectoryValue =
"../../data/Weight";
algorithm->setInputParameter("dataDirectory", &dataDirectoryValue);
// Crossover operator
double crParameter = 1.0;
double fParameter = 0.5;
parameters["CR"] = &crParameter;
parameters["F"] = &fParameter;
crossover = new DifferentialEvolutionCrossover(parameters);
// Mutation operator
parameters.clear();
double probabilityParameter = 1.0/(problem->getNumberOfVariables());
double distributionIndexParameter = 20.0;
parameters["probability"] = &probabilityParameter;
parameters["distributionIndex"] = &distributionIndexParameter;
mutation = new PolynomialMutation(parameters);
// Add the operators to the algorithm
algorithm->addOperator("crossover",crossover);
algorithm->addOperator("mutation",mutation);
// Add the indicator object to the algorithm
//algorithm->setInputParameter("indicators", indicators) ;
// Execute the Algorithm
t_ini = clock();
SolutionSet * population = algorithm->execute();
t_fin = clock();
double secs = (double) (t_fin - t_ini);
secs = secs / CLOCKS_PER_SEC;
// Result messages
cout << "Total execution time: " << secs << "s" << endl;
cout << "Variables values have been written to file VAR" << endl;
population->printVariablesToFile("VAR");
cout << "Objectives values have been written to file FUN" << endl;
population->printObjectivesToFile("FUN");
delete mutation;
delete crossover;
delete population;
delete algorithm;
} // main
TEST_F(ProjectFixture, RubyMeasureRecord_BCLMeasure) {
// Construct problem with RubyMeasure that points to BCLMeasure
Problem problem("Problem",VariableVector(),runmanager::Workflow());
MeasureGroup dvar("Variable",MeasureVector());
problem.push(dvar);
openstudio::path measuresPath = resourcesPath() / toPath("/utilities/BCL/Measures/v2");
openstudio::path dir = measuresPath / toPath("SetWindowToWallRatioByFacade");
ASSERT_TRUE(BCLMeasure::load(dir));
BCLMeasure measure = BCLMeasure::load(dir).get();
RubyMeasure rpert(measure);
dvar.push(rpert);
OSArgument arg = OSArgument::makeDoubleArgument("wwr");
arg.setValue(0.4);
rpert.setArgument(arg);
arg = OSArgument::makeIntegerArgument("typo_arg");
arg.setDefaultValue(1);
rpert.setArgument(arg);
// Serialize to database
{
ProjectDatabase database = getCleanDatabase("RubyMeasureRecord_BCLMeasure");
bool didStartTransaction = database.startTransaction();
EXPECT_TRUE(didStartTransaction);
// Problem Record
ProblemRecord problemRecord = ProblemRecord::factoryFromProblem(problem,database);
database.save();
if (didStartTransaction) {
EXPECT_TRUE(database.commitTransaction());
}
}
// Re-open database, de-serialize, verify that RubyMeasure is intact.
openstudio::path tempDir1 = measuresPath / toPath(toString(createUUID()));
{
ProjectDatabase database = getExistingDatabase("RubyMeasureRecord_BCLMeasure");
ASSERT_EQ(1u,ProblemRecord::getProblemRecords(database).size());
ASSERT_EQ(1u,MeasureGroupRecord::getMeasureGroupRecords(database).size());
EXPECT_EQ(1u,RubyMeasureRecord::getRubyMeasureRecords(database).size());
MeasureRecordVector dprs = MeasureGroupRecord::getMeasureGroupRecords(database)[0].measureRecords(false);
ASSERT_EQ(1u,dprs.size());
ASSERT_TRUE(dprs[0].optionalCast<RubyMeasureRecord>());
RubyMeasureRecord rpr = dprs[0].cast<RubyMeasureRecord>();
RubyMeasure rp = rpr.rubyMeasure();
EXPECT_TRUE(rp.usesBCLMeasure());
EXPECT_TRUE(rp.measure());
EXPECT_EQ(dir,rp.measureDirectory());
EXPECT_EQ(measure.uuid(),rp.measureUUID());
EXPECT_EQ(measure.versionUUID(),rp.measureVersionUUID());
EXPECT_ANY_THROW(rp.perturbationScript());
EXPECT_EQ(2u,rp.arguments().size());
EXPECT_FALSE(rp.hasIncompleteArguments());
// Update measure and save
BCLMeasure newVersion = measure.clone(tempDir1).get();
newVersion.setDescription("Window to wall ratio with sill height configurable.");
newVersion.save();
EXPECT_NE(measure.versionUUID(),newVersion.versionUUID());
OSArgumentVector args;
args.push_back(OSArgument::makeDoubleArgument("wwr"));
args.push_back(OSArgument::makeDoubleArgument("sillHeight"));
Problem problemCopy = ProblemRecord::getProblemRecords(database)[0].problem();
problemCopy.updateMeasure(newVersion,args,false);
bool didStartTransaction = database.startTransaction();
EXPECT_TRUE(didStartTransaction);
// Problem Record
ProblemRecord problemRecord = ProblemRecord::factoryFromProblem(problemCopy,database);
database.save();
if (didStartTransaction) {
EXPECT_TRUE(database.commitTransaction());
}
}
// Re-open database, check that old argument records are gone, check that de-serialized object ok
openstudio::path tempDir2 = measuresPath / toPath(toString(createUUID()));
{
ProjectDatabase database = getExistingDatabase("RubyMeasureRecord_BCLMeasure");
ASSERT_EQ(1u,ProblemRecord::getProblemRecords(database).size());
EXPECT_EQ(1u,MeasureGroupRecord::getMeasureGroupRecords(database).size());
EXPECT_EQ(1u,RubyMeasureRecord::getRubyMeasureRecords(database).size());
EXPECT_EQ(1u,FileReferenceRecord::getFileReferenceRecords(database).size());
EXPECT_EQ(2u,OSArgumentRecord::getOSArgumentRecords(database).size());
Problem problemCopy = ProblemRecord::getProblemRecords(database)[0].problem();
InputVariableVector vars = problemCopy.variables();
ASSERT_FALSE(vars.empty());
ASSERT_TRUE(vars[0].optionalCast<MeasureGroup>());
MeasureVector dps = vars[0].cast<MeasureGroup>().measures(false);
ASSERT_FALSE(dps.empty());
ASSERT_TRUE(dps[0].optionalCast<RubyMeasure>());
RubyMeasure rp = dps[0].cast<RubyMeasure>();
EXPECT_TRUE(rp.usesBCLMeasure());
EXPECT_TRUE(rp.measure());
EXPECT_EQ(tempDir1,rp.measureDirectory());
EXPECT_EQ(measure.uuid(),rp.measureUUID());
EXPECT_NE(measure.versionUUID(),rp.measureVersionUUID());
EXPECT_ANY_THROW(rp.perturbationScript());
ASSERT_EQ(2u,rp.arguments().size());
EXPECT_TRUE(rp.hasIncompleteArguments());
EXPECT_EQ("wwr",rp.arguments()[0].name());
ASSERT_EQ(1u,rp.incompleteArguments().size());
EXPECT_EQ("sillHeight",rp.incompleteArguments()[0].name());
// Set to different measure
BCLMeasure measure2 = measure.clone(tempDir2).get();
measure2.changeUID();
measure2.incrementVersionId();
measure2.save();
measure2 = BCLMeasure::load(tempDir2).get();
EXPECT_NE(measure.uuid(),measure2.uuid());
EXPECT_NE(measure.versionUUID(),measure2.versionUUID());
rp.setMeasure(measure2);
EXPECT_TRUE(rp.isDirty());
EXPECT_TRUE(problemCopy.isDirty());
bool didStartTransaction = database.startTransaction();
EXPECT_TRUE(didStartTransaction);
// Problem Record
ProblemRecord problemRecord = ProblemRecord::factoryFromProblem(problemCopy,database);
database.save();
if (didStartTransaction) {
EXPECT_TRUE(database.commitTransaction());
}
}
// Re-open database, check that old measure and all argument records are gone
{
ProjectDatabase database = getExistingDatabase("RubyMeasureRecord_BCLMeasure");
ASSERT_EQ(1u,ProblemRecord::getProblemRecords(database).size());
EXPECT_EQ(1u,MeasureGroupRecord::getMeasureGroupRecords(database).size());
EXPECT_EQ(1u,RubyMeasureRecord::getRubyMeasureRecords(database).size());
EXPECT_EQ(1u,FileReferenceRecord::getFileReferenceRecords(database).size());
EXPECT_EQ(0u,OSArgumentRecord::getOSArgumentRecords(database).size());
Problem problemCopy = ProblemRecord::getProblemRecords(database)[0].problem();
InputVariableVector vars = problemCopy.variables();
ASSERT_FALSE(vars.empty());
ASSERT_TRUE(vars[0].optionalCast<MeasureGroup>());
MeasureVector dps = vars[0].cast<MeasureGroup>().measures(false);
ASSERT_FALSE(dps.empty());
ASSERT_TRUE(dps[0].optionalCast<RubyMeasure>());
RubyMeasure rp = dps[0].cast<RubyMeasure>();
EXPECT_TRUE(rp.usesBCLMeasure());
EXPECT_TRUE(rp.measure());
EXPECT_EQ(tempDir2,rp.measureDirectory());
EXPECT_NE(measure.uuid(),rp.measureUUID());
EXPECT_NE(measure.versionUUID(),rp.measureVersionUUID());
EXPECT_ANY_THROW(rp.perturbationScript());
ASSERT_EQ(0u,rp.arguments().size());
EXPECT_FALSE(rp.hasIncompleteArguments());
}
boost::filesystem::remove_all(tempDir1);
boost::filesystem::remove_all(tempDir2);
}
