org.apache.ignite.ml.knn

Class KNNModel<L>

java.lang.Object

org.apache.ignite.ml.knn.KNNModel<L>

Type Parameters:

L - Label type.

All Implemented Interfaces:

Serializable, AutoCloseable, IgniteModel<Vector,L>, Model<Vector,L>, SpatialIndex<L>

Direct Known Subclasses:

KNNClassificationModel, KNNRegressionModel

public abstract class KNNModel<L>
extends Object
implements IgniteModel<Vector,L>, SpatialIndex<L>

KNN model build on top of distribtued spatial indices. Be aware that this model is linked with cluster environment it's been built on and can't be saved or used in other places. Under the hood it keeps Dataset that consists of a set of resources allocated across the cluster.

See Also:

Serialized Form

Field Summary

Fields

Modifier and Type	Field and Description
protected DistanceMeasure	distanceMeasure Distance measure.
protected int	k Number of neighbours.
protected boolean	weighted Weighted or not.

Constructor Summary

Constructors

Modifier	Constructor and Description
protected	KNNModel(Dataset<EmptyContext,SpatialIndex<L>> dataset, DistanceMeasure distanceMeasure, int k, boolean weighted) Constructs a new instance of KNN model.

Method Summary

Modifier and Type	Method and Description
void	close()
List<LabeledVector<L>>	findKClosest(int k, Vector pnt) Finds k closest elements to the specified point.

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, combine, toString

Methods inherited from interface org.apache.ignite.ml.inference.Model

predict

Field Detail

distanceMeasure

protected final DistanceMeasure distanceMeasure

Distance measure.

k

protected final int k

Number of neighbours.

weighted

protected final boolean weighted

Weighted or not.

Constructor Detail

KNNModel

protected KNNModel(Dataset<EmptyContext,SpatialIndex<L>> dataset,
                   DistanceMeasure distanceMeasure,
                   int k,
                   boolean weighted)

Constructs a new instance of KNN model.

Parameters:

dataset - Dataset with SpatialIndex as a partition data.

distanceMeasure - Distance measure.

k - Number of neighbours.

weighted - Weighted or not.

Method Detail

findKClosest

public List<LabeledVector<L>> findKClosest(int k,
                                           Vector pnt)

Finds k closest elements to the specified point.

Specified by:

findKClosest in interface SpatialIndex<L>

Parameters:

k - Number of elements to be returned.

pnt - Point to be used to calculate distance to other points.

Returns:

An array of the k closest elements to the specified point.

close

public void close()

Specified by:

close in interface AutoCloseable

Specified by:

close in interface IgniteModel<Vector,L>

Specified by:

close in interface Model<Vector,L>

Specified by:

close in interface SpatialIndex<L>