gradientgmm

Type Members

1. class GradientGaussianMixture extends UpdatableGaussianMixture with Optimizable

   

   Optimizable gradient-based Gaussian Mixture Model See An Alternative to EM for Gaussian Mixture Models: Batch and Stochastic Riemannian Optimization]]

2. class MetricAggregator extends Serializable

   

   Distributed aggregator of necessary statistics

   Distributed aggregator of necessary statistics

   In each worker it computes and aggregates the current batch log-likelihood, and the terms that will later be used to compute the gradient. The class structure is based on Spark's ExpectationSum