org.tribuo.multilabel.sgd.linear

Class LinearSGDOptions

java.lang.Object

org.tribuo.multilabel.sgd.linear.LinearSGDOptions

All Implemented Interfaces:

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

public class LinearSGDOptions
extends Object
implements com.oracle.labs.mlrg.olcut.config.Options

CLI options for training a linear classifier.

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	LinearSGDOptions.LossEnum Available loss types.

Field Summary

Fields

Modifier and Type	Field and Description
int	sgdEpochs
int	sgdLoggingInterval
int	sgdMinibatchSize
LinearSGDOptions.LossEnum	sgdObjective
GradientOptimiserOptions	sgoOptions

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

header

Constructor Summary

Constructors

Constructor and Description
LinearSGDOptions()

Method Summary

Modifier and Type	Method and Description
MultiLabelObjective	getLoss() Returns the loss function specified in the arguments.
LinearSGDTrainer	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

formatUsage, generateTypeDescription, getAllOptions, getEnumConstantString, getOptionFields, getOptions, getOptionsDescription, getOptionUsage, getOptionUsage, getOptionUsage, getUsage

Field Detail

sgoOptions

public GradientOptimiserOptions sgoOptions

sgdEpochs

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

sgdObjective

@Option(longName="sgd-objective",
        usage="Loss function.")
public LinearSGDOptions.LossEnum sgdObjective

sgdLoggingInterval

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

sgdMinibatchSize

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

Constructor Detail

LinearSGDOptions

public LinearSGDOptions()

Method Detail

getLoss

public MultiLabelObjective getLoss()

Returns the loss function specified in the arguments.

Returns:

The loss function.

getTrainer

public LinearSGDTrainer getTrainer()