smile.sequence

Class HMMLabeler<T>

java.lang.Object

smile.sequence.HMMLabeler<T>

All Implemented Interfaces:

java.io.Serializable, SequenceLabeler<T>

public class HMMLabeler<T>
extends java.lang.Object
implements SequenceLabeler<T>

First-order Hidden Markov Model sequence labeler.

See Also:

Serialized Form

Field Summary

Fields

Modifier and Type	Field and Description
HMM	model The HMM model.
java.util.function.ToIntFunction<T>	ordinal The lambda returns the ordinal numbers of symbols.

Constructor Summary

Constructors

Constructor and Description
HMMLabeler(HMM model, java.util.function.ToIntFunction<T> ordinal) Constructor.

Method Summary

Modifier and Type	Method and Description
static <T> HMMLabeler<T>	fit(T[][] observations, int[][] labels, java.util.function.ToIntFunction<T> ordinal) Fits an HMM by maximum likelihood estimation.
double	logp(T[] o) Returns the logarithm probability of an observation sequence.
double	logp(T[] o, int[] s) Returns the log joint probability of an observation sequence along a state sequence.
double	p(T[] o) Returns the probability of an observation sequence.
double	p(T[] o, int[] s) Returns the joint probability of an observation sequence along a state sequence.
int[]	predict(T[] o) Returns the most likely state sequence given the observation sequence by the Viterbi algorithm, which maximizes the probability of P(I | O, HMM).
java.lang.String	toString()
void	update(T[][] observations, int iterations) Updates the HMM by the Baum-Welch algorithm.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, wait, wait, wait

Field Detail

model

public final HMM model

The HMM model.

ordinal

public final java.util.function.ToIntFunction<T> ordinal

The lambda returns the ordinal numbers of symbols.

Constructor Detail

HMMLabeler

public HMMLabeler(HMM model,
                  java.util.function.ToIntFunction<T> ordinal)

Constructor.

Parameters:

model - the HMM model.

ordinal - a lambda returning the ordinal numbers of symbols.

Method Detail

fit

public static <T> HMMLabeler<T> fit(T[][] observations,
                                    int[][] labels,
                                    java.util.function.ToIntFunction<T> ordinal)

Fits an HMM by maximum likelihood estimation.

Parameters:

observations - the observation sequences.

labels - the state labels of observations, of which states take values in [0, p), where p is the number of hidden states.

ordinal - a lambda returning the ordinal numbers of symbols.

update

public void update(T[][] observations,
                   int iterations)

Updates the HMM by the Baum-Welch algorithm.

Parameters:

observations - the training observation sequences.

iterations - the number of iterations to execute.

toString

public java.lang.String toString()

Overrides:

toString in class java.lang.Object

p

public double p(T[] o,
                int[] s)

Returns the joint probability of an observation sequence along a state sequence.

Parameters:

o - an observation sequence.

s - a state sequence.

Returns:

the joint probability P(o, s | H) given the model H.

logp

public double logp(T[] o,
                   int[] s)

Returns the log joint probability of an observation sequence along a state sequence.

Parameters:

o - an observation sequence.

s - a state sequence.

Returns:

the log joint probability P(o, s | H) given the model H.

p

public double p(T[] o)

Returns the probability of an observation sequence.

Parameters:

o - an observation sequence.

Returns:

the probability of this sequence.

logp

public double logp(T[] o)

Returns the logarithm probability of an observation sequence. A scaling procedure is used in order to avoid underflows when computing the probability of long sequences.

Parameters:

o - an observation sequence.

Returns:

the log probability of this sequence.

predict

public int[] predict(T[] o)

Returns the most likely state sequence given the observation sequence by the Viterbi algorithm, which maximizes the probability of P(I | O, HMM). In the calculation, we may get ties. In this case, one of them is chosen randomly.

Specified by:

predict in interface SequenceLabeler<T>

Parameters:

o - an observation sequence.

Returns:

the most likely state sequence.