HiddenMarkovModelTDistribution, TObservationDecode Method (TObservation, Double)

Note: This API is now obsolete.

Calculates the most likely sequence of hidden states that produced the given observation sequence.

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

Syntax

Copy

[ObsoleteAttribute("Please set Algorithm to Viterbi and call Decide() instead.")]
public int[] Decode(
	TObservation[] observations,
	out double logLikelihood
)

<ObsoleteAttribute("Please set Algorithm to Viterbi and call Decide() instead.")>
Public Function Decode ( 
	observations As TObservation(),
	<OutAttribute> ByRef logLikelihood As Double
) As Integer()

Request Example View Source

Parameters

observations: Type: TObservation
A sequence of observations.
logLikelihood: Type: SystemDouble
The log-likelihood along the most likely sequence.

Return Value

Type: Int32
The sequence of states that most likely produced the sequence.

Remarks

Decoding problem. Given the HMM M = (A, B, pi) and the observation sequence O = {o1,o2, ..., oK}, calculate the most likely sequence of hidden states Si that produced this observation sequence O. This can be computed efficiently using the Viterbi algorithm.

Reference

HiddenMarkovModelTDistribution, TObservation Class

Decode Overload

Accord.Statistics.Models.Markov Namespace