smile.classification

Class Maxent.Trainer

java.lang.Object

smile.classification.ClassifierTrainer<int[]>

smile.classification.Maxent.Trainer

Enclosing class:

Maxent

public static class Maxent.Trainer
extends ClassifierTrainer<int[]>

Trainer for maximum entropy classifier.

Constructor Summary

Constructors

Constructor and Description
Trainer(int p) Constructor.

Method Summary

Modifier and Type	Method and Description
Maxent.Trainer	setMaxNumIteration(int maxIter) Sets the maximum number of BFGS stopping iterations.
Maxent.Trainer	setRegularizationFactor(double lambda) Sets the regularization factor.
Maxent.Trainer	setTolerance(double tol) Sets the tolerance for BFGS stopping iterations.
Maxent	train(int[][] 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

Trainer

public Trainer(int p)

Constructor.

Parameters:

p - the dimension of feature space.

Method Detail

setRegularizationFactor

public Maxent.Trainer setRegularizationFactor(double lambda)

Sets the regularization factor. λ > 0 gives a "regularized" estimate of linear weights which often has superior generalization performance, especially when the dimensionality is high.

Parameters:

lambda - regularization factor.

setTolerance

public Maxent.Trainer setTolerance(double tol)

Sets the tolerance for BFGS stopping iterations.

Parameters:

tol - tolerance for stopping iterations.

setMaxNumIteration

public Maxent.Trainer setMaxNumIteration(int maxIter)

Sets the maximum number of BFGS stopping iterations.

Parameters:

maxIter - maximum number of iterations.

train

public Maxent train(int[][] x,
                    int[] y)

Description copied from class: ClassifierTrainer

Learns a classifier with given training data.

Specified by:

train in class ClassifierTrainer<int[]>

Parameters:

x - the training instances.

y - the training labels.

Returns:

a trained classifier.