optimise_prior: optimise_prior

In gtonkinhill/fastbaps: A fast genetic clustering algorithm that approximates a Dirichlet Process Mixture model

optimise_prior

Description

Function to optimise the prior using Bayes Factors and grid search

Usage

optimise_prior(
  sparse.data,
  grid.interval = c(5e-04, 10),
  type = "optimise.baps",
  hc.method = "ward",
  n.cores = 1
)

Arguments

sparse.data	a sparse SNP data object returned from import_fasta_sparse_nt
grid.interval	the upper and lower bound for the hyperparameter to optimise over (default=c(5e-4, 10))
type	one of 'optimise.symmetric','symmetric', 'optimise.baps', 'baps' or 'hc'. (default='optimise.baps')
hc.method	the type of initial hierarchical clustering to use. Can be with 'ward' or 'genie' (default='ward')
n.cores	number of cores to use (currently not implemented)

Value

a sparse.data object with the prior optimised via grid search

Examples

fasta.file.name <- system.file("extdata", "seqs.fa", package = "fastbaps")
sparse.data <- import_fasta_sparse_nt(fasta.file.name)
sparse.data <- optimise_prior(sparse.data)

gtonkinhill/fastbaps documentation built on Sept. 25, 2022, 1:56 p.m.