jmaxent
Class Train

java.lang.Object
  jmaxent.Train

public class Train
extends java.lang.Object

The Class Train.

Field Summary

Model	model The model.
int	numFeatures The num features.
int	numLabels The num labels.

Constructor Summary

Train()
Instantiates a new train.

Method Summary

double	computeLogLiGradient(double[] lambda, double[] gradLogLi, int numIter, java.io.PrintWriter fout) Compute log li gradient.
void	doTrain(java.io.PrintWriter fout) Do train.
void	init() Inits the.
static double	norm(double[] vect) Norm.

Methods inherited from class java.lang.Object

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

Field Detail

model

public Model model

The model.

numLabels

public int numLabels

The num labels.

numFeatures

public int numFeatures

The num features.

Constructor Detail

Train

public Train()

Instantiates a new train.

Method Detail

init

public void init()

Inits the.

norm

public static double norm(double[] vect)

Norm.

Parameters:

vect - the vect

Returns:

the double

doTrain

public void doTrain(java.io.PrintWriter fout)

Do train.

Parameters:

fout - the fout

computeLogLiGradient

public double computeLogLiGradient(double[] lambda,
                                   double[] gradLogLi,
                                   int numIter,
                                   java.io.PrintWriter fout)

Compute log li gradient.

Parameters:

lambda - the lambda

gradLogLi - the grad log li

numIter - the num iter

fout - the fout

Returns:

the double