Slicer< V, L, R > Class Template Reference

#include <slicer.h>

Public Member Functions
	Slicer (int Dim, const V &scale, int kmax=10)
void	run (L &lnprob, std::vector< V > &chain, V xo, R &ran, int step_type, bool verbose=false)
	run the MC chain
size_t	number_evaluations ()
	returns the number of evaluations of the posterior during the last run

Detailed Description

template<typename V, typename L, typename R>
class Slicer< V, L, R >

This is a slice sampler that can be used to do a markov chain without repeated entries in the chain.

types; V is usuatly a std::vector<float> or std::vector<double> L is the functor type used for the likelihood function, this functure should have a opertor()(V p) that will return the log of the likeliwood

Constructor & Destructor Documentation

Slicer()

template<typename V , typename L , typename R >

Slicer< V, L, R >::Slicer ( int Dim, const V & scale, int kmax = 10 )

inline

Parameters

Dim	number of parameters
scale	initial stepsize in parameter space
kmax	maximuma number of attempts made in each step

Member Function Documentation

run()

template<typename V , typename L , typename R >

void Slicer< V, L, R >::run ( L & lnprob, std::vector< V > & chain, V xo, R & ran, int step_type, bool verbose = false )

inline

run the MC chain

Parameters

lnprob	log likelihood function or funtor
chain	will contain the MC chain on return. should be initialized to the desired length
xo	initial point in parameter space
ran	rendom number generator
step_type	0 step_out, !=0 step_double

---

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

• SLsimLib/include/slicer.h