AnyType
dist_norm1::run(AnyType& args) {
return distNorm1(
args[0].getAs<MappedColumnVector>(),
args[1].getAs<MappedColumnVector>()
);
}
int main(int argc, char** argv) {
NormalDistribution<1> distNorm1(5, 5);
std::vector<double> samplesNorm1;
distNorm1.getSamples(samplesNorm1, 1000);
EstimatorML<NormalDistribution<1> > estNorm1;
estNorm1.addPoints(samplesNorm1.begin(), samplesNorm1.end());
std::cout << "Estimation1: " << std::endl << estNorm1 << std::endl;
NormalDistribution<2> distNorm2(Eigen::Matrix<double, 2, 1>(5, 5),
(Eigen::Matrix<double, 2, 2>() << 5.0, 0.5, 0.5, 5.0).finished());
std::vector<Eigen::Matrix<double, 2, 1> > samplesNorm2;
distNorm2.getSamples(samplesNorm2, 1000);
EstimatorML<NormalDistribution<2> > estNorm2;
estNorm2.addPoints(samplesNorm2.begin(), samplesNorm2.end());
std::cout << "Estimation2: " << std::endl << estNorm2 << std::endl;
CategoricalDistribution<4> distCat4(
(Eigen::Matrix<double, 4, 1>() << 0.1, 0.2, 0.2, 0.5).finished());
std::vector<Eigen::Matrix<int, 4, 1> > samplesCat4;
distCat4.getSamples(samplesCat4, 1000);
EstimatorML<CategoricalDistribution<4> > estCat4;
estCat4.addPoints(samplesCat4.begin(), samplesCat4.end());
std::cout << "Estimation3: " << std::endl << estCat4 << std::endl;
MultinomialDistribution<4> distMult4(20,
(Eigen::Matrix<double, 4, 1>() << 0.1, 0.2, 0.2, 0.5).finished());
std::vector<Eigen::Matrix<int, 4, 1> > samplesMult4;
distMult4.getSamples(samplesMult4, 1000);
EstimatorML<MultinomialDistribution<4> > estMult4;
estMult4.addPoints(samplesMult4.begin(), samplesMult4.end());
std::cout << "Estimation4: " << std::endl << estMult4 << std::endl;
ExponentialDistribution distExp(5);
std::vector<double> samplesExp;
distExp.getSamples(samplesExp, 1000);
EstimatorML<ExponentialDistribution> estExp;
estExp.addPoints(samplesExp.begin(), samplesExp.end());
std::cout << "Estimation5: " << std::endl << estExp << std::endl;
GeometricDistribution distGeom(0.7);
std::vector<int> samplesGeom;
distGeom.getSamples(samplesGeom, 1000);
EstimatorML<GeometricDistribution> estGeom;
estGeom.addPoints(samplesGeom.begin(), samplesGeom.end());
std::cout << "Estimation6: " << std::endl << estGeom << std::endl;
PoissonDistribution distPois(5.0);
std::vector<int> samplesPois;
distPois.getSamples(samplesPois, 1000);
EstimatorML<PoissonDistribution> estPois;
estPois.addPoints(samplesPois.begin(), samplesPois.end());
std::cout << "Estimation7: " << std::endl << estPois << std::endl;
LinearRegression<2> distLine(LinearBasisFunction<double, 2>(
Eigen::Matrix<double, 2, 1>(2.0, 2.0)), 2.0);
std::vector<Eigen::Matrix<double, 2, 1> > samplesLine;
for (double x = -10; x < 10; x += 0.01) {
distLine.setBasis((Eigen::Matrix<double, 1, 1>() << x).finished());
samplesLine.push_back(distLine.getSample());
}
EstimatorML<LinearRegression<2> > estLine;
estLine.addPoints(samplesLine.begin(), samplesLine.end());
std::cout << "Estimation8: " << std::endl << estLine << std::endl;
std::vector<NormalDistribution<1> > distributionsNorm1;
distributionsNorm1.push_back(NormalDistribution<1>(0, 1.5));
distributionsNorm1.push_back(NormalDistribution<1>(5, 1.7));
distributionsNorm1.push_back(NormalDistribution<1>(10, 1.2));
distributionsNorm1.push_back(NormalDistribution<1>(-5, 1.3));
distributionsNorm1.push_back(NormalDistribution<1>(-10, 1.5));
MixtureSampleDistribution<NormalDistribution<1>, 5> distMixtNorm1(
distributionsNorm1, CategoricalDistribution<5>(
(Eigen::Matrix<double, 5, 1>() << 0.1, 0.2, 0.2, 0.4, 0.1).finished()));
std::vector<double> samplesMixtNorm1;
distMixtNorm1.getSamples(samplesMixtNorm1, 1000);
std::vector<NormalDistribution<1> > distributionsNorm1Init;
distributionsNorm1Init.push_back(NormalDistribution<1>(1, 2.4));
distributionsNorm1Init.push_back(NormalDistribution<1>(6, 3.1));
distributionsNorm1Init.push_back(NormalDistribution<1>(12, 4.7));
distributionsNorm1Init.push_back(NormalDistribution<1>(-5.5, 2.1));
distributionsNorm1Init.push_back(NormalDistribution<1>(-9.5, 1.4));
MixtureSampleDistribution<NormalDistribution<1>, 5> distMixtNorm1Init(
distributionsNorm1Init, CategoricalDistribution<5>());
EstimatorML<MixtureDistribution<NormalDistribution<1>, 5> >
estMixtNorm1(distMixtNorm1Init);
size_t numIter = estMixtNorm1.addPointsEM(samplesMixtNorm1.begin(),
samplesMixtNorm1.end());
std::cout << "Estimation9: " << std::endl << estMixtNorm1 << std::endl;
std::cout << "Converged in: " << numIter << " iterations" << std::endl;
estMixtNorm1 = EstimatorML<MixtureDistribution<NormalDistribution<1>, 5> >
(distMixtNorm1Init);
numIter = estMixtNorm1.addPointsCEM(samplesMixtNorm1.begin(),
samplesMixtNorm1.end());
std::cout << "Estimation10: " << std::endl << estMixtNorm1 << std::endl;
std::cout << "Converged in: " << numIter << " iterations" << std::endl;
estMixtNorm1 = EstimatorML<MixtureDistribution<NormalDistribution<1>, 5> >
(distMixtNorm1Init);
numIter = estMixtNorm1.addPointsSEM(samplesMixtNorm1.begin(),
samplesMixtNorm1.end());
std::cout << "Estimation11: " << std::endl << estMixtNorm1 << std::endl;
std::cout << "Converged in: " << numIter << " iterations" << std::endl;
std::vector<NormalDistribution<3> > distributionsNorm3;
distributionsNorm3.push_back(NormalDistribution<3>(
Eigen::Matrix<double, 3, 1>(0, 0, 0),
Eigen::Matrix<double, 3, 3>::Identity()));
distributionsNorm3.push_back(NormalDistribution<3>(
Eigen::Matrix<double, 3, 1>(5, 5, 5),
Eigen::Matrix<double, 3, 3>::Identity()));
distributionsNorm3.push_back(NormalDistribution<3>(
Eigen::Matrix<double, 3, 1>(10, 10, 10),
Eigen::Matrix<double, 3, 3>::Identity()));
distributionsNorm3.push_back(NormalDistribution<3>(
Eigen::Matrix<double, 3, 1>(-5, -5, -5),
Eigen::Matrix<double, 3, 3>::Identity()));
distributionsNorm3.push_back(NormalDistribution<3>(
Eigen::Matrix<double, 3, 1>(-10, -10, -10),
Eigen::Matrix<double, 3, 3>::Identity()));
MixtureSampleDistribution<NormalDistribution<3>, 5> distMixtNorm3(
distributionsNorm3, CategoricalDistribution<5>());
std::vector<Eigen::Matrix<double, 3, 1> > samplesMixtNorm3;
distMixtNorm3.getSamples(samplesMixtNorm3, 1000);
std::vector<NormalDistribution<3> > distributionsNorm3Init;
distributionsNorm3Init.push_back(NormalDistribution<3>(
Eigen::Matrix<double, 3, 1>(0.1, 0.2, 0.3)));
distributionsNorm3Init.push_back(NormalDistribution<3>(
Eigen::Matrix<double, 3, 1>(5.5, 6, 5.1)));
distributionsNorm3Init.push_back(NormalDistribution<3>(
Eigen::Matrix<double, 3, 1>(10.5, 11, 12)));
distributionsNorm3Init.push_back(NormalDistribution<3>(
Eigen::Matrix<double, 3, 1>(-5, -5.4, -4.2)));
distributionsNorm3Init.push_back(NormalDistribution<3>(
Eigen::Matrix<double, 3, 1>(-10.5, -10.2, -11)));
MixtureSampleDistribution<NormalDistribution<3>, 5> distMixtNorm3Init(
distributionsNorm3Init, CategoricalDistribution<5>());
EstimatorML<MixtureDistribution<NormalDistribution<3>, 5> > estMixtNorm3(
distMixtNorm3Init);
numIter = estMixtNorm3.addPointsEM(samplesMixtNorm3.begin(),
samplesMixtNorm3.end());
std::cout << "Estimation12: " << std::endl << estMixtNorm3 << std::endl;
std::cout << "Converged in: " << numIter << " iterations" << std::endl;
std::vector<LinearRegression<2> > distLines;
distLines.push_back(LinearRegression<2>(LinearBasisFunction<double, 2>(
Eigen::Matrix<double, 2, 1>(1.0, 1.0)), 1.5));
distLines.push_back(LinearRegression<2>(LinearBasisFunction<double, 2>(
Eigen::Matrix<double, 2, 1>(2.0, 2.0)), 1.2));
distLines.push_back(LinearRegression<2>(LinearBasisFunction<double, 2>(
Eigen::Matrix<double, 2, 1>(-1.0, -1.0)), 2.1));
distLines.push_back(LinearRegression<2>(LinearBasisFunction<double, 2>(
Eigen::Matrix<double, 2, 1>(-2.0, -2.0)), 2.5));
MixtureSampleDistribution<LinearRegression<2>, 4> distMixtLines(distLines,
CategoricalDistribution<4>());
std::vector<Eigen::Matrix<double, 2, 1> > samplesMixtLines;
for (size_t i = 0; i < 100; ++i)
for (double x = -10; x < 10; x += 0.1) {
for (size_t j = 0; j < distLines.size(); ++j)
distLines[j].setBasis((Eigen::Matrix<double, 1, 1>() << x).finished());
distMixtLines.setCompDistributions(distLines);
samplesMixtLines.push_back(distMixtLines.getSample());
}
EstimatorML<MixtureDistribution<LinearRegression<2>, 4> > estMixtLines(
distMixtLines);
numIter = estMixtLines.addPointsEM(samplesMixtLines.begin(),
samplesMixtLines.end());
std::cout << "Estimation13: " << std::endl << estMixtLines << std::endl;
std::cout << "Converged in: " << numIter << " iterations" << std::endl;
return 0;
}
