R/mallet_state_io.R

In mallet: An R Wrapper for the Java Mallet Topic Modeling Toolkit

#' Save a Mallet state to file
#'
#' @description
#' This function writes a current sampling state of mallet to file.
#' The state contain hyperparameters \eqn{\alpha} and \eqn{\beta} together with topic indicators.
#'
#' The state file can be read into R using the function
#'
#' @param topic.model
#' A \code{cc.mallet.topics.RTopicModel} object created by \code{\link{MalletLDA}}.
#' @param state.file File path (.gz format) to store the mallet state file to.
#'
#' @export
save.mallet.state <- function(topic.model, state.file){
  checkmate::assert_path_for_output(state.file)
  checkmate::assert_character(state.file, pattern = "\\.gz$")
  checkmate::assert_class(topic.model, "jobjRef")
  checkmate::assert_string(utils::capture.output(print(topic.model)), "Java-Object{cc.mallet.topics.RTopicModel@[a-zA-z0-9]+}")

  topic.model$writeState(state.file)
}

#' Load a Mallet state into Mallet
#'
#' @description
#' This reads writes a current sampling state of mallet to file. The state contain
#' hyperparameters \eqn{\alpha} and \eqn{\beta} together with topic indicators.
#'
#' @param topic.model
#' A \code{cc.mallet.topics.RTopicModel} object created by \code{\link{MalletLDA}}.
#' @param state.file File path to store the mallet state file to.
#'
#' @returns a java \code{cc.mallet.topics.RTopicModel} object
#'
#' @export
load.mallet.state <- function(topic.model, state.file){
  checkmate::assert_file_exists(state.file)
  checkmate::assert_class(topic.model, "jobjRef")
  checkmate::assert_string(utils::capture.output(print(topic.model)), "Java-Object{cc.mallet.topics.RTopicModel@[a-zA-z0-9]+}")

  jFile <- rJava::.jnew("java/io/File", state.file)
  topic.model$initializeFromState(jFile)
}