Class PerceptronTrainer

Namespace
NOpenNLP.Tools.Ml.Perceptron
Assembly
NOpenNLP.Tools.dll

Trains models using the perceptron algorithm. Each outcome is represented as a binary perceptron classifier. This supports standard (integer) weighting as well as average weighting as described in: Discriminative Training Methods for Hidden Markov Models: Theory and Experiments with the Perceptron Algorithm. Michael Collins, EMNLP 2002.

public class PerceptronTrainer : AbstractEventTrainer, IEventTrainer
Inheritance
PerceptronTrainer
Implements
Inherited Members

Constructors

PerceptronTrainer() 

public PerceptronTrainer()

PerceptronTrainer(TrainingParameters) 

public PerceptronTrainer(TrainingParameters parameters)

Parameters

parameters TrainingParameters

Fields

PERCEPTRON_VALUE 

public const string PERCEPTRON_VALUE = "PERCEPTRON"

Field Value

string

TOLERANCE_DEFAULT 

public const double TOLERANCE_DEFAULT = 1E-05

Field Value

double

Properties

IsSortAndMerge

Whether the data indexer should sort and merge the indexed events.

public override bool IsSortAndMerge { get; }

Property Value

bool

SkippedAveraging

Enables skipped averaging; this flag changes the standard averaging to special averaging instead.

If we are doing averaging, and the current iteration is one of the first 20 or it is a perfect square, then update the summed parameters.

The reason we don't take all of them is that the parameters change less toward the end of training, so they drown out the contributions of the more volatile early iterations. The use of perfect squares allows us to sample from successively farther apart iterations.

public virtual bool SkippedAveraging { set; }

Property Value

bool

StepSizeDecrease

Enables and sets step size decrease. The step size is decreased every iteration by the specified value, in percent.

public virtual double StepSizeDecrease { set; }

Property Value

double

Tolerance

Specifies the tolerance. If the change in training set accuracy is less than this, stop iterating.

public virtual double Tolerance { set; }

Property Value

double

Methods

DoTrain(IDataIndexer)

Trains a model from the given, already indexed, training data.

public override IMaxentModel DoTrain(IDataIndexer indexer)

Parameters

indexer IDataIndexer

Returns

IMaxentModel

TrainModel(int, IDataIndexer, int) 

public virtual AbstractModel TrainModel(int iterations, IDataIndexer di, int cutoff)

Parameters

iterations int
di IDataIndexer
cutoff int

Returns

AbstractModel

TrainModel(int, IDataIndexer, int, bool) 

public virtual AbstractModel TrainModel(int iterations, IDataIndexer di, int cutoff, bool useAverage)

Parameters

iterations int
di IDataIndexer
cutoff int
useAverage bool

Returns

AbstractModel

Validate()

Checks the parameters. If a subclass overrides this, it should call the base implementation.

public override void Validate()

Exceptions

ArgumentException

Thrown if a parameter is not valid.

Edit this page