ForwardBackwardAlgorithmLogBackward Method (HiddenMarkovModel, Int32, Double)

Computes Backward probabilities for a given hidden Markov model and a set of observations (no scaling).

Namespace: Accord.Statistics.Models.Markov
Assembly: Accord.Statistics (in Accord.Statistics.dll) Version: 3.8.0

Syntax

Copy

public static double[,] LogBackward(
	HiddenMarkovModel model,
	int[] observations,
	out double logLikelihood
)

Public Shared Function LogBackward ( 
	model As HiddenMarkovModel,
	observations As Integer(),
	<OutAttribute> ByRef logLikelihood As Double
) As Double(,)

Request Example View Source

Parameters

model: Type: Accord.Statistics.Models.MarkovHiddenMarkovModel
observations: Type: SystemInt32
logLikelihood: Type: SystemDouble

Return Value

Type: Double

Reference

ForwardBackwardAlgorithm Class

LogBackward Overload

Accord.Statistics.Models.Markov Namespace

Send comments on this topic to feedback@accord-framework.net