Class MultivariateGaussian

Object
org.apache.spark.mllib.stat.distribution.MultivariateGaussian
All Implemented Interfaces:
Serializable

public class MultivariateGaussian extends Object implements Serializable
This class provides basic functionality for a Multivariate Gaussian (Normal) Distribution. In the event that the covariance matrix is singular, the density will be computed in a reduced dimensional subspace under which the distribution is supported. (see Degenerate case in Multivariate normal distribution (Wikipedia))

param: mu The mean vector of the distribution param: sigma The covariance matrix of the distribution

See Also:
  • Constructor Details

    • MultivariateGaussian

      public MultivariateGaussian(Vector mu, Matrix sigma)
  • Method Details

    • logpdf

      public double logpdf(Vector x)
      Returns the log-density of this multivariate Gaussian at given point, x
      Parameters:
      x - (undocumented)
      Returns:
      (undocumented)
    • mu

      public Vector mu()
    • pdf

      public double pdf(Vector x)
      Returns density of this multivariate Gaussian at given point, x
      Parameters:
      x - (undocumented)
      Returns:
      (undocumented)
    • sigma

      public Matrix sigma()