Package org.tribuo.regression.sgd.fm

Class TrainTest.FMRegressionOptions

java.lang.Object

org.tribuo.regression.sgd.fm.TrainTest.FMRegressionOptions

All Implemented Interfaces:

com.oracle.labs.mlrg.olcut.config.Options

Enclosing class:

TrainTest

public static class TrainTest.FMRegressionOptions
extends Object
implements com.oracle.labs.mlrg.olcut.config.Options

Command line options.

Field Summary

Fields

Modifier and Type	Field	Description
int	epochs	Number of SGD epochs.
int	factorSize	Factor size.
DataOptions	general	The dataset loading options.
GradientOptimiserOptions	gradientOptions	Gradient descent options.
int	loggingInterval	Log the objective after n examples.
TrainTest.LossEnum	loss	Loss function.
int	minibatchSize	Minibatch size.
boolean	standardise	Standardise the output regressors before model fitting.
double	variance	Variance of the initialization gaussian.

Fields inherited from interface com.oracle.labs.mlrg.olcut.config.Options

header

Constructor Summary

Constructors

Constructor	Description
FMRegressionOptions()

Method Summary

Modifier and Type	Method	Description
String	getOptionsDescription()

Methods inherited from class java.lang.Object

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

Field Details

general

public DataOptions general

The dataset loading options.

gradientOptions

public GradientOptimiserOptions gradientOptions

Gradient descent options.

epochs

@Option(charName='i',
        longName="epochs",
        usage="Number of SGD epochs.")
public int epochs

Number of SGD epochs. Defaults to 5.

loss

@Option(charName='o',
        longName="objective",
        usage="Loss function.")
public TrainTest.LossEnum loss

Loss function. Defaults to SQUARED.

loggingInterval

@Option(charName='p',
        longName="logging-interval",
        usage="Log the objective after <int> examples.")
public int loggingInterval

Log the objective after n examples. Defaults to 100.

minibatchSize

@Option(charName='z',
        longName="minibatch-size",
        usage="Minibatch size.")
public int minibatchSize

Minibatch size. Defaults to 1.

factorSize

@Option(charName='d',
        longName="factor-size",
        usage="Factor size.")
public int factorSize

Factor size.

variance

@Option(longName="variance",
        usage="Variance of the initialization gaussian.")
public double variance

Variance of the initialization gaussian.

standardise

@Option(longName="standardise",
        usage="Standardise the output regressors before model fitting.")
public boolean standardise

Standardise the output regressors before model fitting.

Constructor Details

FMRegressionOptions

public FMRegressionOptions()

Method Details

getOptionsDescription

public String getOptionsDescription()

Specified by:

getOptionsDescription in interface com.oracle.labs.mlrg.olcut.config.Options