Represents a Markov Chain Monte Carlo sampler that returns samples from a distribution. More...

#include <mcmc.hpp>

Public Member Functions
	Sampler (const MCMCSettings &s, const DBSettings &d, uint stateDim, volatile bool &interrupted)
	Create a new sampler. More...

template<class AsyncPolicy , class PropFn >
void	run (AsyncPolicy &policy, const std::vector< Eigen::VectorXd > &initialStates, PropFn &propFn, uint numSeconds)
	Run the sampler for a period of time. More...

ChainArray	chains ()
	Get the MCMC chain array. More...

const ChainArray &	chains () const
	Get the MCMC chain array. More...

Detailed Description

Represents a Markov Chain Monte Carlo sampler that returns samples from a distribution.

Constructor & Destructor Documentation

stateline::mcmc::Sampler::Sampler	(	const MCMCSettings &	s,
		const DBSettings &	d,
		uint	stateDim,
		volatile bool &	interrupted
	)

inline

Create a new sampler.

Parameters

s	Settings to tune various parameters of the MCMC.
d	Settings for configuring the database for storing samples.
stateDim	The number of dimensions in each sample.
interrupted	A flag used to monitor whether the sampler has been interrupted.

ChainArray stateline::mcmc::Sampler::chains ( )

inline

Get the MCMC chain array.

const ChainArray& stateline::mcmc::Sampler::chains ( ) const

inline

Get the MCMC chain array.

template<class AsyncPolicy , class PropFn >

void stateline::mcmc::Sampler::run	(	AsyncPolicy &	policy,
		const std::vector< Eigen::VectorXd > &	initialStates,
		PropFn &	propFn,
		uint	numSeconds
	)

inline

Run the sampler for a period of time.

Parameters

policy	Async policy to evaluate states.
initialStates	Initial chain states. Ignored if recovering.
propFn	The proposal function.
numSeconds	The number of seconds to run the MCMC for.

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