vmcmc::Algorithm Class Reference

Abstract base class for the core MCMC sampling algorithms. More...

#include <algorithm.hpp>

Inheritance diagram for vmcmc::Algorithm:

Collaboration diagram for vmcmc::Algorithm:

Public Types
using	DefaultCallable = std::function< double(const std::vector< double > &)>

Public Member Functions
void	SetParameterConfig (const ParameterConfig &paramConfig)
	Set the parameter configuration. More...
const ParameterConfig &	GetParameterConfig () const
void	SetPrior (DefaultCallable prior)
template<size_t NParams, typename CallableT >
void	SetPrior (CallableT prior)
void	SetLikelihood (DefaultCallable likelihood)
template<size_t NParams, typename CallableT >
void	SetLikelihood (CallableT likelihood)
void	SetNegLogLikelihood (DefaultCallable negLoglikelihood)
template<size_t NParams, typename CallableT >
void	SetNegLogLikelihood (CallableT negLoglikelihood)
void	SetTotalLength (size_t length)
size_t	GetTotalLength () const
template<typename WriterT , typename... ArgsT>
void	AddWriter (ArgsT &&...args)
	Add an output writer by specifying its type and passing constructor arguments. More...
void	AddWriter (std::shared_ptr< Writer > writer)
double	EvaluatePrior (const std::vector< double > &paramValues) const
	Evaluate the prior for the given parameter values. More...
double	EvaluateLikelihood (const std::vector< double > &paramValues) const
	Evaluate the target function likelihood for the given parameter values. More...
double	EvaluateNegLogLikelihood (const std::vector< double > &paramValues) const
	Evaluate the target function -log(likelihood) for the given parameter values. More...
bool	Evaluate (Sample &sample) const
	Evalutate the target function prior, likelihood and -log(likelihood) at the position defined by the sample, and update the sample accordingly. More...
void	Run ()
	Start sampling!
virtual void	Initialize ()
virtual void	Advance (size_t nSteps=1)=0
virtual void	Finalize ()
virtual size_t	NumberOfChains ()=0
virtual const Chain &	GetChain (size_t cIndex=0)=0
ChainSetStatistics &	GetStatistics ()
const ChainSetStatistics &	GetStatistics () const

Protected Attributes
ParameterConfig	fParameterConfig
DefaultCallable	fPrior
DefaultCallable	fLikelihood
DefaultCallable	fNegLogLikelihood
size_t	fTotalLength
size_t	fCycleLength
std::vector< std::shared_ptr< Writer > >	fWriters
ChainSetStatistics	fStatistics

Detailed Description

Abstract base class for the core MCMC sampling algorithms.

This base class models the sampling algorithm of an MCMC, advancing to a new state in the parameter space, accepting or rejecting, and saving that state (subsequently sampling the parameter space).

Ideally, this interface should cover single chain algorithms, as well as multi-chain algorithms (like DREAM).

Member Function Documentation

template<typename WriterT , typename... ArgsT>

void vmcmc::Algorithm::AddWriter ( ArgsT &&...  args )

inline

Add an output writer by specifying its type and passing constructor arguments.

Parameters

args

bool vmcmc::Algorithm::Evaluate

(

Sample &

sample

)

const

Evalutate the target function prior, likelihood and -log(likelihood) at the position defined by the sample, and update the sample accordingly.

Parameters

sample

Returns

False if the likelihood was not evaluated (e.g. due to a zero prior).

double vmcmc::Algorithm::EvaluateLikelihood

(

const std::vector< double > &

paramValues

)

const

Evaluate the target function likelihood for the given parameter values.

Parameters

paramValues

Returns

The likelihood value.

double vmcmc::Algorithm::EvaluateNegLogLikelihood

(

const std::vector< double > &

paramValues

)

const

Evaluate the target function -log(likelihood) for the given parameter values.

Parameters

paramValues

Returns

The negative logarithm (natural base) of the likelihood.

double vmcmc::Algorithm::EvaluatePrior

(

const std::vector< double > &

paramValues

)

const

Evaluate the prior for the given parameter values.

Parameters

paramValues

Returns

The prior value. Returns 1.0 if no priors were defined.

void vmcmc::Algorithm::SetParameterConfig

(

const ParameterConfig &

paramConfig

)

Set the parameter configuration.

This is mandatory for every possible implementations of this class. From the configuration, the start points and proposal functions of a sampler a initialized. The number of parameters defines, with how many arguments the target likelihood function is evaluated.

Parameters

paramConfig

---

The documentation for this class was generated from the following files:

• vmcmc/algorithm.hpp
• vmcmc/algorithm.cpp