JSD.matrix: Jensen-Shannon Divergence (for more than two distributions)

In clintpgeorge/clda: Approximate Inference Algorithms for the Compound Latent Dirichlet Allocation Model

Description

For more than two distributions, we need a function to compute the Entropy H(v)

Usage

JSD.matrix(w, M)

Arguments

w	A vector of weights which should sum up to 1
M	a matrix with the input distributions as columns

Note

Adapted from the discussion http://stackoverflow.com/questions/11226627/jensen-shannon-divergence-in-r

See Also

Other utils: H, JSD, KLD, sample_antoniak, sample_dirichlet, sample_multinomial

clintpgeorge/clda documentation built on May 13, 2019, 8 p.m.