lti::hmmOnlineClassifier::parameters Class Reference

the parameters for the class hmmOnlineClassifier More...

#include <ltiHmmOnlineClassifier.h>

Inheritance diagram for lti::hmmOnlineClassifier::parameters:

[legend]

Collaboration diagram for lti::hmmOnlineClassifier::parameters:

[legend]

List of all members.

Public Member Functions
	parameters ()
	parameters (const parameters &other)
virtual	~parameters ()
const char *	getTypeName () const
parameters &	copy (const parameters &other)
parameters &	operator= (const parameters &other)
virtual classifier::parameters *	clone () const
virtual bool	write (ioHandler &handler, const bool complete=true) const
virtual bool	read (ioHandler &handler, const bool complete=true)
Public Attributes
float	pruningThreshold
int	maxActiveHypotheses
int	numberOfBuckets
int	automaticTraceBack

---

Detailed Description

the parameters for the class hmmOnlineClassifier

---

Constructor & Destructor Documentation

lti::hmmOnlineClassifier::parameters::parameters

(

)

default constructor

Reimplemented from lti::hmmClassifier::parameters.

lti::hmmOnlineClassifier::parameters::parameters

(

const parameters &

other

)

copy constructor

Parameters:

other

the parameters object to be copied

Reimplemented from lti::hmmClassifier::parameters.

virtual lti::hmmOnlineClassifier::parameters::~parameters

(

)

[virtual]

destructor

Reimplemented from lti::hmmClassifier::parameters.

---

Member Function Documentation

virtual classifier::parameters* lti::hmmOnlineClassifier::parameters::clone

(

)

const [virtual]

returns a pointer to a clone of the parameters

Reimplemented from lti::hmmClassifier::parameters.

parameters& lti::hmmOnlineClassifier::parameters::copy

(

const parameters &

other

)

copy the contents of a parameters object

Parameters:

other

the parameters object to be copied

Returns:

a reference to this parameters object

Reimplemented from lti::hmmClassifier::parameters.

const char* lti::hmmOnlineClassifier::parameters::getTypeName

(

)

const [virtual]

returns name of this type

Reimplemented from lti::hmmClassifier::parameters.

parameters& lti::hmmOnlineClassifier::parameters::operator=

(

const parameters &

other

)

copy the contents of a parameters object

Parameters:

other

the parameters object to be copied

Returns:

a reference to this parameters object

Reimplemented from lti::hmmClassifier::parameters.

virtual bool lti::hmmOnlineClassifier::parameters::read	(	ioHandler &	handler,
		const bool	complete = true
	)			[virtual]

read the parameters from the given ioHandler

Parameters:

	handler	the ioHandler to be used
	complete	if true (the default) the enclosing begin/end will be also written, otherwise only the data block will be written.

Returns:

true if write was successful

Reimplemented from lti::hmmClassifier::parameters.

virtual bool lti::hmmOnlineClassifier::parameters::write	(	ioHandler &	handler,
		const bool	complete = true
	)			const [virtual]

write the parameters in the given ioHandler

Parameters:

	handler	the ioHandler to be used
	complete	if true (the default) the enclosing begin/end will be also written, otherwise only the data block will be written.

Returns:

true if write was successful

Reimplemented from lti::hmmClassifier::parameters.

---

Member Data Documentation

int lti::hmmOnlineClassifier::parameters::automaticTraceBack

after the given number of timesteps, a partial trace back is performed and the calculated values are saved internally.

this is done to keep the trace back field small and the scores low. a value of 0 disables automatic trace back. default is 0.

int lti::hmmOnlineClassifier::parameters::maxActiveHypotheses

describes the maximum number of active hypotheses (histogram pruning).

if more hypotheses (i.e. paths within a model) are active, the worst hypotheses are thrown away. for value 0, no histogram pruning is performed. this value should be adjusted depending on the number of models. default is 1000.

int lti::hmmOnlineClassifier::parameters::numberOfBuckets

defines the number of buckets for the bucket-sort-algorithm used in histogram pruning.

when performing histogram pruning, one bucket-sort-run is performed and the hypotheses in the buckets for the lowest scores are kept, so that there are at most maxActiveHypotheses active. the more buckets there are, the closer the number of surviving hypotheses will get to the given maximum count. (the number of buckets has no influence on run time!) default is 100.

float lti::hmmOnlineClassifier::parameters::pruningThreshold

additive constant used for pruning (beam search).

all hypotheses that have a score greater than minScore + pruningThreshold are thrown away. For values lower than 0, no beam search is performed. default is 1000, but this is no generally reasonable value. the threshold depends strongly on the number of features and the trained data.

---

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

• ltiHmmOnlineClassifier.h