Class FMClassificationOptions

java.lang.Object

org.tribuo.classification.sgd.fm.FMClassificationOptions

All Implemented Interfaces:

com.oracle.labs.mlrg.olcut.config.Options, ClassificationOptions<FMClassificationTrainer>

public class FMClassificationOptions
extends Object
implements ClassificationOptions<FMClassificationTrainer>

CLI options for training a factorization machine classifier.

Nested Class Summary

Nested Classes

Modifier and Type	Class	Description
static enum	FMClassificationOptions.LossEnum	Available loss types.

Field Summary

Fields

Modifier and Type	Field	Description
int	fmEpochs	Number of SGD epochs.
int	fmFactorSize	Factor size.
int	fmLoggingInterval	Log the objective after n examples.
int	fmMinibatchSize	Minibatch size.
FMClassificationOptions.LossEnum	fmObjective	Loss function.
double	fmVariance	Variance of the initialization gaussian.
GradientOptimiserOptions	sgoOptions	The gradient descent algorithm options.

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

header

Constructor Summary

Constructors

Constructor	Description
FMClassificationOptions()

Method Summary

Modifier and Type	Method	Description
LabelObjective	getLoss()	Returns the loss function specified in the arguments.
FMClassificationTrainer	getTrainer()

Methods inherited from class java.lang.Object

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

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

getOptionsDescription

Field Details

sgoOptions

public GradientOptimiserOptions sgoOptions

The gradient descent algorithm options.

fmEpochs

@Option(longName="fm-epochs",
        usage="Number of SGD epochs.")
public int fmEpochs

Number of SGD epochs.

fmObjective

@Option(longName="fm-objective",
        usage="Loss function.")
public FMClassificationOptions.LossEnum fmObjective

Loss function.

fmLoggingInterval

@Option(longName="fm-logging-interval",
        usage="Log the objective after <int> examples.")
public int fmLoggingInterval

Log the objective after n examples.

fmMinibatchSize

@Option(longName="fm-minibatch-size",
        usage="Minibatch size.")
public int fmMinibatchSize

Minibatch size.

fmFactorSize

@Option(longName="fm-factor-size",
        usage="Factor size.")
public int fmFactorSize

Factor size.

fmVariance

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

Variance of the initialization gaussian.

Constructor Details

FMClassificationOptions

public FMClassificationOptions()

Method Details

getLoss

public LabelObjective getLoss()

Returns the loss function specified in the arguments.

Returns:

The loss function.

getTrainer

public FMClassificationTrainer getTrainer()

Specified by:

getTrainer in interface ClassificationOptions<FMClassificationTrainer>