Abstract Class representing all _FULLY_ Discrete Conditional PDF's. More...

#include <discreteconditionalpdf.h>

Inheritance diagram for BFL::DiscreteConditionalPdf:

Public Member Functions
virtual DiscreteConditionalPdf *	Clone () const
	Clone function.
	DiscreteConditionalPdf (int num_states=1, int num_conditional_arguments=1, int cond_arg_dimensions[]=NULL)
	Constructor.
	DiscreteConditionalPdf (const DiscreteConditionalPdf &pdf)
	Copy constructor.
unsigned int	NumStatesGet () const
	Get the number of discrete states.
Probability	ProbabilityGet (const int &input) const
	Get the probability of a certain argument.
void	ProbabilitySet (const double &prob, const int &input, const std::vector< int > &condargs) const
	Set the probability (Typical for discrete Pdf's)
virtual bool	SampleFrom (Sample< int > &one_sample, int method, void *args) const
	Draw 1 sample from the Pdf:
virtual bool	SampleFrom (vector< Sample< int > > &list_samples, int num_samples, int method, void *args) const
virtual	~DiscreteConditionalPdf ()
	Destructor.
Protected Member Functions
int	IndexGet (const int &input, const std::vector< int > &condargs) const
	Get the correct index in the row of doubles (double * probability)
Protected Attributes
int *	_cond_arg_dims_p
	"Possible discrete states" of all the conditional arguments
unsigned int	_num_states
	number of discrete states
double *	_probability_p
	Pointer to the probability values.
std::vector< double >	_probs
int	_total_dimension
	Total dimension of the likelihoodtable.
std::vector< double >	_valuelist

Detailed Description

Abstract Class representing all _FULLY_ Discrete Conditional PDF's.

This class inherits only from ConditionalPdf (not from DiscretePdf, avoiding a circular class structure

            ------
            |    |
            ------
           /      \
      -----        -----
      |   |        |   |
      -----        -----
           \      /
            ------
	    |    |
	    ------

Todo:: Check if this is the best way to implement this.

Note:: that the name of this class could be better chosen. Something like Discrete-DiscreteConditionalPdf would maybe be more clear (???), but quite long...

See also:: ConditionalPdf

Definition at line 53 of file discreteconditionalpdf.h.

Constructor & Destructor Documentation

BFL::DiscreteConditionalPdf::DiscreteConditionalPdf	(	int	num_states = `1`,
		int	num_conditional_arguments = `1`,
		int	cond_arg_dimensions[] = `NULL`
	)

Constructor.

Precondition:: The number of elements of cond_arg_dimensions should be equal to num_conditional_arguments, otherwise -> Segfaults

Parameters:

num_states	int representing the number of possible states
num_conditional_arguments	the number of arguments behind the \|
cond_arg_dimensions[]	possible number of states of the different conditional arguments

See also:: ConditionalPdf

Todo:: Get cleaner api and implementation

Definition at line 28 of file discreteconditionalpdf.cpp.

BFL::DiscreteConditionalPdf::DiscreteConditionalPdf ( const DiscreteConditionalPdf & pdf )

Copy constructor.

Definition at line 61 of file discreteconditionalpdf.cpp.

BFL::DiscreteConditionalPdf::~DiscreteConditionalPdf ( ) [virtual]

Destructor.

Definition at line 51 of file discreteconditionalpdf.cpp.

Member Function Documentation

DiscreteConditionalPdf * BFL::DiscreteConditionalPdf::Clone ( ) const [virtual]

Clone function.

Reimplemented from BFL::ConditionalPdf< int, int >.

Definition at line 84 of file discreteconditionalpdf.cpp.

int BFL::DiscreteConditionalPdf::IndexGet	(	const int &	input,
		const std::vector< int > &	condargs
	)		const `[protected]`

Get the correct index in the row of doubles (double * probability)

Definition at line 96 of file discreteconditionalpdf.cpp.

unsigned int BFL::DiscreteConditionalPdf::NumStatesGet ( ) const

Get the number of discrete states.

Definition at line 90 of file discreteconditionalpdf.cpp.

Probability BFL::DiscreteConditionalPdf::ProbabilityGet ( const int & input ) const [virtual]

Get the probability of a certain argument.

Parameters:

input T argument of the Pdf

Returns:: the probability value of the argument

Reimplemented from BFL::BFL::Pdf< int >.

Definition at line 119 of file discreteconditionalpdf.cpp.

void BFL::DiscreteConditionalPdf::ProbabilitySet	(	const double &	prob,
		const int &	input,
		const std::vector< int > &	condargs
	)		const

Set the probability (Typical for discrete Pdf's)

Definition at line 127 of file discreteconditionalpdf.cpp.

bool BFL::DiscreteConditionalPdf::SampleFrom	(	Sample< int > &	one_sample,
		int	method,
		void *	args
	)		const `[virtual]`

Draw 1 sample from the Pdf:

There's no need to create a list for only 1 sample!

Parameters:

one_sample	sample that will contain result of sampling
method	Sampling method to be used. Each sampling method is currently represented by a #define statement, eg. #define BOXMULLER 1
args	Pointer to a struct representing extra sample arguments