Package org.tribuo.common.sgd

Class AbstractSGDModel<T extends Output<T>>

java.lang.Object

org.tribuo.Model<T>

org.tribuo.common.sgd.AbstractSGDModel<T>

All Implemented Interfaces:

com.oracle.labs.mlrg.olcut.provenance.Provenancable<ModelProvenance>, Serializable

Direct Known Subclasses:

AbstractFMModel, AbstractLinearSGDModel

public abstract class AbstractSGDModel<T extends Output<T>>
extends Model<T>

A model trained using SGD.

See:

 Bottou L.
 "Large-Scale Machine Learning with Stochastic Gradient Descent"
 Proceedings of COMPSTAT, 2010.
 

See Also:

• Serialized Form

Nested Class Summary

Nested Classes

Modifier and Type	Class	Description
protected static final class	AbstractSGDModel.PredAndActive	A nominal tuple used to capture the prediction and the number of active features used by the model.

Field Summary

Fields

Modifier and Type	Field	Description
protected boolean	addBias
protected FeedForwardParameters	modelParameters	The weights for this model.

Fields inherited from class org.tribuo.Model

ALL_OUTPUTS, BIAS_FEATURE, featureIDMap, generatesProbabilities, name, outputIDInfo, provenance, provenanceOutput

Constructor Summary

Constructors

Modifier	Constructor	Description
protected	AbstractSGDModel(String name, ModelProvenance provenance, ImmutableFeatureMap featureIDMap, ImmutableOutputInfo<T> outputIDInfo, FeedForwardParameters weights, boolean generatesProbabilities, boolean addBias)	Constructs a linear model trained via SGD.

Method Summary

Modifier and Type	Method	Description
FeedForwardParameters	getModelParameters()	Returns a copy of the model parameters.
protected AbstractSGDModel.PredAndActive	predictSingle(Example<T> example)	Generates the dense vector prediction from the supplied example.

Methods inherited from class org.tribuo.Model

castModel, copy, copy, generatesProbabilities, getExcuse, getExcuses, getFeatureIDMap, getName, getOutputIDInfo, getProvenance, getTopFeatures, innerPredict, predict, predict, predict, setName, toString, validate

Methods inherited from class java.lang.Object

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

Field Details

modelParameters

protected FeedForwardParameters modelParameters

The weights for this model.

addBias

protected boolean addBias

Constructor Details

AbstractSGDModel

protected AbstractSGDModel(String name,
 ModelProvenance provenance,
 ImmutableFeatureMap featureIDMap,
 ImmutableOutputInfo<T> outputIDInfo,
 FeedForwardParameters weights,
 boolean generatesProbabilities,
 boolean addBias)

Constructs a linear model trained via SGD.

Parameters:

name - The model name.

provenance - The model provenance.

featureIDMap - The feature domain.

outputIDInfo - The output domain.

weights - The model weights.

generatesProbabilities - Does this model generate probabilities?

addBias - Should the model add a bias feature to the feature vector?

Method Details

predictSingle

protected AbstractSGDModel.PredAndActive predictSingle(Example<T> example)

Generates the dense vector prediction from the supplied example.

Parameters:

example - The example to use for prediction.

Returns:

The prediction and the number of features involved.

getModelParameters

public FeedForwardParameters getModelParameters()

Returns a copy of the model parameters.

Returns:

A copy of the model parameters.