de.jungblut.online.regression

Class RegressionLearner

java.lang.Object

de.jungblut.online.ml.AbstractOnlineLearner<M>

de.jungblut.online.ml.AbstractMinimizingOnlineLearner<RegressionModel>

de.jungblut.online.regression.RegressionLearner

All Implemented Interfaces:

OnlineLearner<RegressionModel>

public class RegressionLearner
extends AbstractMinimizingOnlineLearner<RegressionModel>

A regression learner that learns weights on a stream, given an optimization objective (e.g. log loss). This learner outputs a RegressionModel that can be used in a RegressionClassifier.

Author:

thomas.jungblut

Field Summary

Fields inherited from class de.jungblut.online.ml.AbstractMinimizingOnlineLearner

minimizer, numPasses, random, sparseWeights

Fields inherited from class de.jungblut.online.ml.AbstractOnlineLearner

featureDimension, numOutcomeClasses, outcomeDimension, streamSupplier, verbose

Constructor Summary

Constructors

Constructor and Description
RegressionLearner(StochasticMinimizer minimizer, de.jungblut.math.activation.ActivationFunction activationFunction, de.jungblut.math.loss.LossFunction lossFunction)

Method Summary

Modifier and Type	Method and Description
RegressionModel	createModel(de.jungblut.math.DoubleVector weights) Creates a model with the given minimized weights.
protected de.jungblut.math.minimize.CostGradientTuple	observeExample(de.jungblut.online.ml.FeatureOutcomePair next, de.jungblut.math.DoubleVector weights) Observes the next example.

Methods inherited from class de.jungblut.online.ml.AbstractMinimizingOnlineLearner

observeExampleSafe, randomInitialize, setNumPasses, setRandom, train, useSparseWeights

Methods inherited from class de.jungblut.online.ml.AbstractOnlineLearner

init, peekDimensions, setVerbose, verbose

Methods inherited from class java.lang.Object

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

Constructor Detail

RegressionLearner

public RegressionLearner(StochasticMinimizer minimizer,
                         de.jungblut.math.activation.ActivationFunction activationFunction,
                         de.jungblut.math.loss.LossFunction lossFunction)

Method Detail

observeExample

protected de.jungblut.math.minimize.CostGradientTuple observeExample(de.jungblut.online.ml.FeatureOutcomePair next,
                                                                     de.jungblut.math.DoubleVector weights)

Description copied from class: AbstractMinimizingOnlineLearner

Observes the next example.

Specified by:

observeExample in class AbstractMinimizingOnlineLearner<RegressionModel>

Parameters:

next - the next feature/outcome pair.

weights - the current weights.

Returns:

a cost gradient tuple that can be used for minimization.

createModel

public RegressionModel createModel(de.jungblut.math.DoubleVector weights)

Description copied from class: AbstractMinimizingOnlineLearner

Creates a model with the given minimized weights.

Specified by:

createModel in class AbstractMinimizingOnlineLearner<RegressionModel>

Parameters:

weights - the learned weights.

Returns:

a model that describes the weights.