transformedWeights

A square matrix of pairwise similarilities between items

A function that maps pairwise similarities. Default is the identity
function.

A numeric value close to 0 to give some nonzero weight to pairs of
items with 0 or 1 pairwise similarity

Enumerate Transformed Weights for Choosing i and j Non-Uniformly

Markov chain Monte Carlo samplers for posterior simulations of conjugate Bayesian nonparametric
mixture models. Functionality is provided for Gibbs sampling as in Algorithm 3 of Neal (2000)
<DOI:10.1080/10618600.2000.10474879>, restricted Gibbs merge-split sampling as described in Jain & Neal
(2004) <DOI:10.1198/1061860043001>, and sequentially-allocated merge-split sampling <DOI:10.1080/00949655.2021.1998502>, as well as
summary and utility functions.

David B Dahl

transformedWeights: Enumerate Transformed Weights for Choosing i and j Non-Uniformly

Description

Usage

Arguments