R/dataset_ag_news.R

In textdata: Download and Load Various Text Datasets

#' AG's News Topic Classification Dataset
#'
#' The AG's news topic classification dataset is constructed by choosing 4
#' largest classes from the original corpus. Each class contains 30,000 training
#' samples and 1,900 testing samples. The total number of training samples is
#' 120,000 and testing 7,600.

#' Version 3, Updated 09/09/2015
#'
#' The classes in this dataset are
#'
#' \itemize{
#' \item World
#' \item Sports
#' \item Business
#' \item Sci/Tech
#' }
#'
#' @inheritParams lexicon_afinn
#' @param split Character. Return training ("train") data or testing ("test")
#'     data. Defaults to "train".
#' @return A tibble with 120,000 or 30,000 rows for "train" and "test"
#'     respectively and 3 variables:
#' \describe{
#'   \item{class}{Character, denoting new class}
#'   \item{title}{Character, title of article}
#'   \item{description}{Character, description of article}
#' }
#' @source \url{http://groups.di.unipi.it/~gulli/AG_corpus_of_news_articles.html}
#' @source \url{https://github.com/srhrshr/torchDatasets/raw/master/dbpedia_csv.tar.gz}
#' @keywords datasets
#' @family topic
#' @export
#' @examples
#' \dontrun{
#' dataset_ag_news()
#'
#' # Custom directory
#' dataset_ag_news(dir = "data/")
#'
#' # Deleting dataset
#' dataset_ag_news(delete = TRUE)
#'
#' # Returning filepath of data
#' dataset_ag_news(return_path = TRUE)
#'
#' # Access both training and testing dataset
#' train <- dataset_ag_news(split = "train")
#' test <- dataset_ag_news(split = "test")
#' }
#'
#' @importFrom fs file_exists dir_exists dir_create path
#' @importFrom readr read_rds
#' @importFrom utils menu
dataset_ag_news <- function(dir = NULL, split = c("train", "test"),
                            delete = FALSE, return_path = FALSE,
                            clean = FALSE, manual_download = FALSE) {
  all_files <- paste0("ag_news_", c("train", "test"), ".rds")
  split <- match.arg(split)
  name <- paste0("ag_news_", split, ".rds")
  load_dataset(
    data_name = "ag_news", name = name, dir = dir,
    delete = delete, return_path = return_path, clean = clean,
    clean_manual = all_files,
    manual_download = manual_download
  )
}

#' @importFrom utils download.file
download_ag_news <- function(folder_path) {
  file_path_test <- path(folder_path, "ag_news_test.csv")
  file_path_train <- path(folder_path, "ag_news_train.csv")

  if (file_exists(file_path_test) & file_exists(file_path_train)) {
    return(invisible())
  }

  download.file(
    url = "https://raw.githubusercontent.com/mhjabreel/CharCnn_Keras/master/data/ag_news_csv/test.csv",
    destfile = file_path_test
  )
  download.file(
    url = "https://raw.githubusercontent.com/mhjabreel/CharCnn_Keras/master/data/ag_news_csv/train.csv",
    destfile = file_path_train
  )
}

#' @importFrom readr read_tsv write_rds cols col_character col_double
#' @importFrom tibble tibble
process_ag_news <- function(folder_path, name_path) {
  file_path_test <- path(folder_path, "ag_news_test.csv")
  file_path_train <- path(folder_path, "ag_news_train.csv")

  data_test <- read_csv(file_path_test,
    col_names = c("class", "title", "description"),
    col_types = cols(
      class = col_double(),
      title = col_character(),
      description = col_character()
    )
  )
  data_train <- read_csv(file_path_train,
    col_names = c("class", "title", "description"),
    col_types = cols(
      class = col_double(),
      title = col_character(),
      description = col_character()
    )
  )

  classes <- c("World", "Sports", "Business", "Sci/Tech")

  data_test$class <- classes[data_test$class]
  data_train$class <- classes[data_train$class]

  write_rds(data_test, path(folder_path, "ag_news_test.rds"))
  write_rds(data_train, path(folder_path, "ag_news_train.rds"))
}