smile.classification

Class FLD.Trainer

java.lang.Object

smile.classification.ClassifierTrainer<double[]>

smile.classification.FLD.Trainer

Enclosing class:

FLD

public static class FLD.Trainer
extends ClassifierTrainer<double[]>

Trainer for Fisher's linear discriminant.

Constructor Summary

Constructors

Constructor and Description
FLD.Trainer() Constructor.

Method Summary

Methods

Modifier and Type	Method and Description
void	setDimension(int L) Sets the dimensionality of mapped space.
void	setTolerance(double tol) Sets covariance matrix singular tolerance.
FLD	train(double[][] x, int[] y) Learns a classifier with given training data.

Methods inherited from class smile.classification.ClassifierTrainer

setAttributes

Methods inherited from class java.lang.Object

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

Constructor Detail

FLD.Trainer

public FLD.Trainer()

Constructor. The dimensionality of mapped space will be k - 1, where k is the number of classes of data. The default tolerance to covariance matrix singularity is 1E-4.

Method Detail

setDimension

public void setDimension(int L)

Sets the dimensionality of mapped space.

Parameters:

L - the dimensionality of mapped space.

setTolerance

public void setTolerance(double tol)

Sets covariance matrix singular tolerance.

Parameters:

tol - a tolerance to decide if a covariance matrix is singular. The trainer will reject variables whose variance is less than tol².

train

public FLD train(double[][] x,
        int[] y)

Description copied from class: ClassifierTrainer

Learns a classifier with given training data.

Specified by:

train in class ClassifierTrainer<double[]>

Parameters:

x - the training instances.

y - the training labels.

Returns:

a trained classifier.