Implements a class for accumulating conditional first- and second-order sufficient statistics. More...

Inheritance diagram for drwnCondSuffStats:

Public Member Functions
	drwnCondSuffStats (int n=1, int k=2, drwnPairSuffStatsType pairStats=DRWN_PSS_FULL)
	construct a sufficient statistic object of dimensionality `n` with `k` conditional states

	drwnCondSuffStats (const drwnCondSuffStats &condStats)
	copy constructor

void	clear ()
	clear the conditional sufficient statistics

void	clear (int n, int k, drwnPairSuffStatsType pairStats=DRWN_PSS_FULL)
	clear the conditional sufficient statistics and re-initialize to dimensionality `n` with `k` conditional states

int	size () const
	return the number of dimensions

int	states () const
	return the number of conditional states

double	count () const
	return the total weight of samples accumulated

double	count (int k) const
	return the weight of samples accumulated for the `k-th` state

drwnSuffStats const &	suffStats (int k) const
	return the sufficient statistics for the `k-th` state

const char *	type () const
	returns object type as a string (e.g., Foo::type() { return "Foo"; })

drwnCondSuffStats *	clone () const
	returns a copy of the class usually implemented as virtual Foo* clone() { return new Foo(*this); }

bool	save (drwnXMLNode &xml) const
	write object to XML node (see also write)

bool	load (drwnXMLNode &xml)
	read object from XML node (see also read)

void	accumulate (const vector< double > &x, int y)
	add a single sample to the sufficient statistics for state `y`

void	accumulate (const vector< double > &x, int y, double w)
	add a single weighted sample to the sufficient statistics for state `y`

void	accumulate (const vector< vector< double > > &x, int y, double w=1.0)
	add a number of samples to the sufficient statistics for state `y`

void	accumulate (const vector< vector< double > > &x, const vector< int > &y)
	add a number of samples to the sufficient statistics for states `y`

void	accumulate (const vector< vector< double > > &x, const vector< int > &y, const vector< double > &w)
	add a number of weighted samples to the sufficient statistics for states `y`

void	accumulate (const drwnSuffStats &stats, int y, double w=1.0)
	add previously accumulated sufficient statistics to the sufficient statistics for state `y`

void	subtract (const vector< double > &x, int y)
	remove a single sample from the sufficient statistics for state `y`

void	subtract (const vector< double > &x, int y, double w)
	remove a single weighted sample from the sufficient statistics for state `y`

void	subtract (const vector< vector< double > > &x, int y, double w=1.0)
	remove a number of samples from the sufficient statistics for state `y`

void	subtract (const vector< vector< double > > &x, const vector< int > &y)
	remove a number of samples from the sufficient statistics for states `y`

void	subtract (const vector< vector< double > > &x, const vector< int > &y, const vector< double > &w)
	remove a number of weighted samples from the sufficient statistics for states `y`

void	subtract (const drwnSuffStats &stats, int y, double w=1.0)
	remove previously accumulated sufficient statistics from the sufficient statistics for state `y`

void	redistribute (int y, int k)
	redistribute the counts from state `y` to state `k`

const drwnSuffStats &	operator[] (unsigned k) const
	copy constructor

Public Member Functions inherited from drwnWriteable
bool	write (const char *filename) const
	write object to file (calls save)

bool	read (const char *filename)
	read object from file (calls load)

void	dump () const
	print object's current state to standard output (for debugging)

Detailed Description

Implements a class for accumulating conditional first- and second-order sufficient statistics.

Calling the function accumulate member function will perform the following updates

count = count + $\sum_{i : y_i = k} w_i$
sum = sum + $\sum_{i : y_i = k} w_i x_i$
sum2 = sum2 + $\sum_{i : y_i = k} w_i x_i x_i^T$

Calling the function subtract member function will perform the following updates

count = count - $\sum_{i : y_i = k} w_i$
sum = sum - $\sum_{i : y_i = k} w_i x_i$
sum2 = sum2 - $\sum_{i : y_i = k} w_i x_i x_i^T$ The statistic count must remain non-negative.

See Also: drwnSuffStats, drwnML Tutorial

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

drwnSuffStats.h
drwnSuffStats.cpp

Public Member Functions

Detailed Description