#include <AdaptiveMetropolisProposal.hpp>
Collaboration diagram for hops::AdaptiveMetropolisProposal< MatrixType, VectorType >:
Public Types | |
| using | StateType = VectorType |
Public Member Functions | |
| AdaptiveMetropolisProposal (MatrixType A, VectorType b, StateType currentState, typename MatrixType::Scalar stepSize=1, typename MatrixType::Scalar eps=1.e-3, unsigned long warmUp=100) | |
| Constructs the Adaptive Metropolis m_proposal mechanism (Haario et al. 2001) on polytope defined as Ax<b. More... | |
| void | propose (RandomNumberGenerator &randomNumberGenerator) |
| void | acceptProposal () |
| StateType | getState () const |
| StateType | getProposal () const |
| void | setState (StateType newState) |
| void | setStepSize (typename MatrixType::Scalar stepSize) |
| MatrixType::Scalar | getStepSize () const |
| MatrixType::Scalar | computeLogAcceptanceProbability () |
| std::string | getName () |
Member Typedef Documentation
◆ StateType
template<typename MatrixType , typename VectorType >
| using hops::AdaptiveMetropolisProposal< MatrixType, VectorType >::StateType = VectorType |
Constructor & Destructor Documentation
◆ AdaptiveMetropolisProposal()
template<typename MatrixType , typename VectorType >
| hops::AdaptiveMetropolisProposal< MatrixType, VectorType >::AdaptiveMetropolisProposal | ( | MatrixType | A, |
| VectorType | b, | ||
| StateType | currentState, | ||
| typename MatrixType::Scalar | stepSize = 1, |
||
| typename MatrixType::Scalar | eps = 1.e-3, |
||
| unsigned long | warmUp = 100 |
||
| ) |
Constructs the Adaptive Metropolis m_proposal mechanism (Haario et al. 2001) on polytope defined as Ax<b.
- Parameters
-
A b currentState stepSize The stepsize by which the trained Gaussian m_proposal distribution is scaled. eps Scaling factor of the maximum volume ellipsoid, which is added to the covariance to preserve (numerical) irreducibility. warmUp Number of warm up samples during which the maximum volume ellipsoid is used as covariance of the m_proposal distribution. After the warm up, the adaptive covariance is used.
Member Function Documentation
◆ acceptProposal()
template<typename MatrixType , typename VectorType >
| void hops::AdaptiveMetropolisProposal< MatrixType, VectorType >::acceptProposal |
◆ computeLogAcceptanceProbability()
template<typename MatrixType , typename VectorType >
| MatrixType::Scalar hops::AdaptiveMetropolisProposal< MatrixType, VectorType >::computeLogAcceptanceProbability |
◆ getName()
template<typename MatrixType , typename VectorType >
| std::string hops::AdaptiveMetropolisProposal< MatrixType, VectorType >::getName |
◆ getProposal()
template<typename MatrixType , typename VectorType >
| AdaptiveMetropolisProposal< MatrixType, VectorType >::StateType hops::AdaptiveMetropolisProposal< MatrixType, VectorType >::getProposal |
◆ getState()
template<typename MatrixType , typename VectorType >
| AdaptiveMetropolisProposal< MatrixType, VectorType >::StateType hops::AdaptiveMetropolisProposal< MatrixType, VectorType >::getState |
◆ getStepSize()
template<typename MatrixType , typename VectorType >
| MatrixType::Scalar hops::AdaptiveMetropolisProposal< MatrixType, VectorType >::getStepSize |
◆ propose()
template<typename MatrixType , typename VectorType >
| void hops::AdaptiveMetropolisProposal< MatrixType, VectorType >::propose | ( | RandomNumberGenerator & | randomNumberGenerator | ) |
◆ setState()
template<typename MatrixType , typename VectorType >
| void hops::AdaptiveMetropolisProposal< MatrixType, VectorType >::setState | ( | StateType | newState | ) |
◆ setStepSize()
template<typename MatrixType , typename VectorType >
| void hops::AdaptiveMetropolisProposal< MatrixType, VectorType >::setStepSize | ( | typename MatrixType::Scalar | stepSize | ) |
The documentation for this class was generated from the following file:
