Classification of samples

Data Structures
struct	akClassHyp
	Classification hypothesis. More...
Typedefs
typedef void(	akClassifyCallback )(void *udata, akProb prob_class, akProb prob_lexicon, akProb prob_sample, size_t class, const akCSeq *cs)
	Callback function for ak_classify.
Functions
void	ak_classifier_free (akClassifier *cls)
	Frees memory.
akClassifier *	ak_classifier_new (const akCModel *cmodel, const akLexicon *lexicon, const akPriors *priors, akClassifyCallback *callback, void *udata, char **err)
	Gets a Classifier.
unsigned int	ak_classify (akClassifier *cls, const akFea *fea, akClassHyp *hyps, const unsigned int N)
	Classifies a sample.

---

Typedef Documentation

typedef void( akClassifyCallback)(void *udata,akProb prob_class,akProb prob_lexicon,akProb prob_sample,size_t class,const akCSeq *cs)

Callback function for ak_classify.

This callback is used to capture probability for every character sequence.

---

Function Documentation

void ak_classifier_free

(

akClassifier *

cls

)

Frees memory.

Frees the memory allocated for the classifier.

Parameters:

cls	The classifier.

akClassifier* ak_classifier_new	(	const akCModel *	cmodel,
		const akLexicon *	lexicon,
		const akPriors *	priors,
		akClassifyCallback *	callback,
		void *	udata,
		char **	err
	)

Gets a Classifier.

This function creates a classifier used to classify samples. Currently supported types are: Bernoulli, DGaussian, Mixture and TiedStates.

Parameters:

cmodel	The character models used to classify.
lexicon	The lexicon used to classify.
priors	The prior probabilities used to classify.
callback	A callback function used to capture the calculated probabilities during the classification process. It can be NULL.
err	Pointer to string variable. If not NULL an error message is allocated in the variable in case of error.
udata	User data passed to the callback function.

Returns:

Classifier or NULL in case of error.

unsigned int ak_classify	(	akClassifier *	cls,
		const akFea *	fea,
		akClassHyp *	hyps,
		const unsigned int	N
	)

Classifies a sample.

This function classifies the given sample, feature vector sequence, according to the next expression: where is the given sample and is a valid pair of character sequence and class. The probability is approximated using the Viterbi algorithm. This function can be used to obtain the N most probable pairs or hypothesis. The variable hyps must contain the enough memory to allocate N hypothesis.

Parameters:

cls	The classifier.
fea	A feature vector manager, which must contain a sequence of feature vectors of an adequate type for the cmodel.
hyps	An array with enough memory to allocate the required hypothesis.
N	Number of required hypothesis.

Returns:

The number of returned hypothesis.