Package smile.classification

Class ClassLabels

java.lang.Object

smile.classification.ClassLabels

All Implemented Interfaces:

Serializable

public class ClassLabels
extends Object
implements Serializable

Map arbitrary class labels to [0, k), where k is the number of classes.

See Also:

• Serialized Form

Field Summary

Fields

Modifier and Type	Field	Description
final IntSet	classes	The class labels.
final int	k	The number of classes.
final int[]	ni	The number of samples per classes.
final double[]	priori	The estimated priori probabilities.
final int[]	y	The sample class id in [0, k).

Constructor Summary

Constructors

Constructor	Description
ClassLabels(int k, int[] y, IntSet classes)	Constructor.

Method Summary

Modifier and Type	Method	Description
static ClassLabels	fit(int[] y)	Fits the class label mapping.
static ClassLabels	fit(Dataset<?,Integer> data)	Fits the class label mapping.
static ClassLabels	fit(BaseVector<?,?,?> response)	Fits the class label mapping.
int[]	indexOf(int[] y)	Maps the class labels to index.
NominalScale	scale()	Returns the nominal scale of the class labels.

Methods inherited from class java.lang.Object

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

Field Details

k

public final int k

The number of classes.

classes

public final IntSet classes

The class labels.

y

public final int[] y

The sample class id in [0, k).

ni

public final int[] ni

The number of samples per classes.

priori

public final double[] priori

The estimated priori probabilities.

Constructor Details

ClassLabels

public ClassLabels(int k,
 int[] y,
 IntSet classes)

Constructor.

Parameters:

k - The number of classes.

y - The sample class id in [0, k).

classes - the class label encoder.

Method Details

scale

public NominalScale scale()

Returns the nominal scale of the class labels.

Returns:

the nominal scale of the class labels.

indexOf

public int[] indexOf(int[] y)

Maps the class labels to index.

Parameters:

y - the sample labels.

Returns:

the indices of labels.

fit

public static ClassLabels fit(Dataset<?,Integer> data)

Fits the class label mapping.

Parameters:

data - the sample instances.

Returns:

the class label mapping.

fit

public static ClassLabels fit(int[] y)

Fits the class label mapping.

Parameters:

y - the sample labels.

Returns:

the class label mapping.

fit

public static ClassLabels fit(BaseVector<?,?,?> response)

Fits the class label mapping.

Parameters:

response - the sample labels.

Returns:

the class label mapping.