gen_plsv: Generate from the PLSV model

In NathanWycoff/iPLSV:

Description

Generate from the PLSV model

Usage

gen_plsv(K, V, M, N.mu, P, eta, gamma, beta)

Arguments

K	The number of topics, an integer scalar.
V	The number of unique words, an integer scalar.
M	The number of documents, an integer scalar.
N.mu	The average number of words per document.
P	The dimensionality of the embedding space, an integer, usually 2.
eta	The exchangible dirichlet prior on words in a topic.
beta	The precision for topic locations, a positive scalar.
gama	The precision for document locations, a positive scalar.

Value

A list containing the generated data, in particular, PHI is the topic by word matrix (rows sum to 1), PSI is a topic location matrix, docs is a list of generated documents, each entry denotes index of vocab, THETA is the document locations, and Z is a list of matrices, with a 1 denoting which topic the word belonged to.

NathanWycoff/iPLSV documentation built on May 16, 2019, 11:10 p.m.