smile.classification

Class RDA.Trainer

java.lang.Object

smile.classification.ClassifierTrainer<double[]>

smile.classification.RDA.Trainer

Enclosing class:

RDA

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

Trainer for regularized discriminant analysis.

Constructor Summary

Constructors

Constructor and Description
RDA.Trainer(double alpha) Constructor.

Method Summary

Methods

Modifier and Type	Method and Description
void	setPriori(double[] priori) Sets a priori probabilities of each class.
void	setTolerance(double tol) Sets covariance matrix singular tolerance.
RDA	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

RDA.Trainer

public RDA.Trainer(double alpha)

Constructor. The default tolerance to covariance matrix singularity is 1E-4.

Parameters:

alpha - regularization factor in [0, 1] allows a continuum of models between LDA and QDA.

Method Detail

setPriori

public void setPriori(double[] priori)

Sets a priori probabilities of each class.

Parameters:

priori - a priori probabilities of each class.

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 RDA 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.