Package marytts.machinelearning

Machine learning classes for K-Means clustering, Gaussian Mixture Models, and manual data generation for testing purposes.

See:
Description

Interface Summary

Discretizer

Class Summary

Cluster	Implements a cluster center that has a mean vector and a covariance matrix (and its inverse)
ClusteredDataGenerator	Generates clustered data for testing machine learning algorithms
ContextualGMMParams	Wrapper for contextual parameters for GMM training - includes various phone identity or class based groups
GaussianComponent	Implements a single Gaussian component with a mean vector and a covariance matrix It also computes terms for pdf computation out of this Gaussian component once the mean and covariance is specified
GMM	Wrapper for a Gaussian Mixture Model
GMMClassifier	TO DO: Implement a GMM based classifier that takes as input several GMMs and data and outputs the probability of each GMM generating the data, the most likely GMM, etc
GmmDiscretizer	This discretizes values according to a gaussian mixture model (gmm).
GMMTrainer	Expectation-Maximization (EM) based GMM training Reference: A.
GMMTrainerParams	Wrapper class for GMM training parameters
KMeansClusteringTrainer	K-Means clustering training algorithm Reference: J.
KMeansClusteringTrainerParams	Wrapper class for K-Means clustering training parameters
PolynomialCluster	Implements a cluster center that has a mean
PolynomialHierarchicalClusteringTrainer	Hierarchical clustering training algorithm Reference: Stephen C.
PolynomialKMeansClusteringTrainer	K-Means clustering training algorithm Reference: J.
SFFS	Sequential Floating Forward Search(SFFS) for selection of features Ref: Pudil, P., J.
SoP	Contains the coefficients and factors of an equation of the form: if interceptTterm = TRUE solution = coeffs[0] + coeffs[1]*factors[0] + coeffs[2]*factors[1] + ...

Package marytts.machinelearning Description

Machine learning classes for K-Means clustering, Gaussian Mixture Models, and manual data generation for testing purposes.