org.apache.ignite.ml.naivebayes.discrete

Class DiscreteNaiveBayesModel

java.lang.Object

org.apache.ignite.ml.naivebayes.discrete.DiscreteNaiveBayesModel

All Implemented Interfaces:

Serializable, AutoCloseable, DeployableObject, Exportable<DiscreteNaiveBayesModel>, IgniteModel<Vector,Double>, Model<Vector,Double>, BayesModel<DiscreteNaiveBayesModel,Vector,Double>

public class DiscreteNaiveBayesModel
extends Object
implements BayesModel<DiscreteNaiveBayesModel,Vector,Double>, DeployableObject

Discrete naive Bayes model which predicts result value y belongs to a class C_k, k in [0..K] as p(C_k,y) =x_1*p_k1^x *...*x_i*p_ki^x_i. Where x_i is a discrete feature, p_ki is a prior probability probability of class p(x|C_k). Returns the number of the most possible class.

See Also:

Serialized Form

Constructor Summary

Constructors

Constructor and Description
DiscreteNaiveBayesModel(double[][][] probabilities, double[] clsProbabilities, double[] labels, double[][] bucketThresholds, DiscreteNaiveBayesSumsHolder sumsHolder)

Method Summary

Modifier and Type	Method and Description
double[][]	getBucketThresholds() A getter for bucketThresholds.
double[]	getClsProbabilities() A getter for clsProbabilities.
List<Object>	getDependencies() Returns dependencies of this object that can be object with class defined by client side and unknown for server.
double[]	getLabels() A getter for labels.
double[][][]	getProbabilities() A getter for probabilities.
DiscreteNaiveBayesSumsHolder	getSumsHolder() A getter for sumsHolder.
Double	predict(Vector vector) Make a prediction for the specified input arguments.
double[]	probabilityPowers(Vector vector) Returns an array where the index correapons a label, and value corresponds log(probability) to be this label.
<P> void	saveModel(Exporter<DiscreteNaiveBayesModel,P> exporter, P path) Save model by the given path.

Methods inherited from class java.lang.Object

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

Methods inherited from interface org.apache.ignite.ml.IgniteModel

andBefore, andThen, andThen, close, combine, toString

Constructor Detail

DiscreteNaiveBayesModel

public DiscreteNaiveBayesModel(double[][][] probabilities,
                               double[] clsProbabilities,
                               double[] labels,
                               double[][] bucketThresholds,
                               DiscreteNaiveBayesSumsHolder sumsHolder)

Parameters:

probabilities - Probabilities of features for classes.

clsProbabilities - Prior probabilities for classes.

labels - Labels.

bucketThresholds - The threshold to convert a feature to a binary value.

sumsHolder - Amount values which are abouve the threshold per label.

Method Detail

saveModel

public <P> void saveModel(Exporter<DiscreteNaiveBayesModel,P> exporter,
                          P path)

Save model by the given path.

Specified by:

saveModel in interface Exportable<DiscreteNaiveBayesModel>

Parameters:

exporter - Exporter.

path - Path to saved model.

predict

public Double predict(Vector vector)

Description copied from interface: Model

Make a prediction for the specified input arguments.

Specified by:

predict in interface Model<Vector,Double>

Parameters:

vector - features vector.

Returns:

A label with max probability.

probabilityPowers

public double[] probabilityPowers(Vector vector)

Returns an array where the index correapons a label, and value corresponds log(probability) to be this label. The prior probabilities are not count.

Specified by:

probabilityPowers in interface BayesModel<DiscreteNaiveBayesModel,Vector,Double>

getProbabilities

public double[][][] getProbabilities()

A getter for probabilities.

getClsProbabilities

public double[] getClsProbabilities()

A getter for clsProbabilities.

getBucketThresholds

public double[][] getBucketThresholds()

A getter for bucketThresholds.

getLabels

public double[] getLabels()

A getter for labels.

getSumsHolder

public DiscreteNaiveBayesSumsHolder getSumsHolder()

A getter for sumsHolder.

getDependencies

public List<Object> getDependencies()

Returns dependencies of this object that can be object with class defined by client side and unknown for server.

Specified by:

getDependencies in interface DeployableObject