Class: GmmDecoder

GmmDecoder

GMM decoder
Loads a model trained by the XMM library and processes an input stream of float vectors in real-time. If the model was trained for regression, outputs an estimation of the associated process.

Constructor

new GmmDecoder(windowSizeopt)

Parameters:
Name Type Attributes Default Description
windowSize Number <optional>
1

Size of the likelihood smoothing window.

Source:

Methods

filter(observation, resultsCallbackopt) → {gmmResults}

The decoding function.

Parameters:
Name Type Attributes Default Description
observation Array

An input float vector to be estimated.

resultsCallback gmmResultsCallback <optional>
null

The callback handling the estimation results.

Source:
Returns:
Type
gmmResults

getLikeliestLabel() → {String}

Get the currently estimated likeliest label.

Source:
Returns:
Type
String

getLikelihoodWindow() → {Number}

Get the likelihood smoothing window size.

Source:
Returns:
Type
Number

getModel() → {xmmGmmModel}

Get the actual XMM GMM model.

Source:
Returns:
Type
xmmGmmModel

getNumberOfClasses() → {Number}

Get the total number of classes the model was trained with.

Source:
Returns:
Type
Number

getRegressionVectorSize() → {Number}

Get the output dimension of the model (size of a regression vector).

Source:
Returns:
Type
Number

setLikelihoodWindow(newWindowSize)

Set the likelihood smoothing window size.

Parameters:
Name Type Description
newWindowSize Number

the new window size.

Source:

setModel(model)

Set the actual XMM GMM model.

Parameters:
Name Type Description
model xmmGmmModel
Source: