plda: Penalized Latent Dirichlet Allocation

In wuxiaotiankevin/pLDA: The pLDA Package

Description

The main function that runs penalized LDA.

Usage

plda(x, k, lambda, warm.start.seed = 2017, verbose = FALSE,
  warm.start.lda = NULL, warm.start.log.beta = NULL, pdata = NULL)

Arguments

x	Input matrix. A matrix of positive integers where rows represents cells (documents) and column represents genes (words).
k	Number of topics.
lambda	Shrinkage parameter.
warm.start.seed	Seed of LDA estimation used as pLDA warm start. pLDA uses parameters estimated from LDA as a warm start to boost speed.
verbose	Default FALSE. Print progress.
warm.start.lda	Default NULL. LDA object for warm start.
warm.start.log.beta	Default NULL. log beta matrix for warm start.
pdata	Default NULL. Table of phenotypic data.

Value

plda() returns a list of the penalized LDA output. logProbW is the log of topic by gene matrix beta. gamma is the cell by topic matrix of topic frequencies for each cell. genes are gene names extracted from the column name of input count matrix.

Examples

## Not run: 
plda(x=cell_by_gene_expr_matrix, k=10, lambda=5)

## End(Not run)

wuxiaotiankevin/pLDA documentation built on Nov. 11, 2019, 11:01 p.m.