X-Git-Url: https://git.auder.net/?p=epclust.git;a=blobdiff_plain;f=epclust%2FR%2Fde_serialize.R;h=f28038b773b5ab995be33a70eb7271822177ca8c;hp=a49990bfad358b16a169d66fca8cbaa9d32a7770;hb=3c5a4b0880db63367a474a568e1322b3999932fe;hpb=40f12a2f66d06fd77183ea02b996f5c66f90761c diff --git a/epclust/R/de_serialize.R b/epclust/R/de_serialize.R index a49990b..f28038b 100644 --- a/epclust/R/de_serialize.R +++ b/epclust/R/de_serialize.R @@ -1,28 +1,28 @@ -#' @name de_serialize -#' @rdname de_serialize -#' @aliases binarize binarizeTransform getDataInFile +#' (De)Serialization of a [big]matrix or data stream #' -#' @title (De)Serialization of a [big]matrix or data stream +#' \code{binarize()} serializes a matrix or CSV file with minimal overhead, into a +#' binary file. \code{getDataInFile()} achieves the inverse task: she retrieves (ASCII) +#' data rows from indices in the binary file. Finally, \code{binarizeTransform()} +#' serialize transformations of all data chunks. To use it a data-retrieval function +#' must be provided -- thus \code{binarize} will most likely be used first +#' (and then a function defined to seek in generated binary file) #' -#' @description \code{binarize()} serializes a matrix or CSV file with minimal overhead, -#' into a binary file. \code{getDataInFile()} achieves the inverse task: she retrieves -#' (ASCII) data rows from indices in the binary file. Finally, -#' \code{binarizeTransform()} serialize transformations of all data chunks; to use it, -#' a data-retrieval function must be provided, thus \code{binarize} will most likely be -#' used first (and then a function defined to seek in generated binary file) -#' -#' @param data_ascii Either a matrix or CSV file, with items in rows -#' @param indices Indices of the lines to retrieve -#' @param data_bin_file Name of binary file on output (\code{binarize}) -#' or input (\code{getDataInFile}) +#' @param data_ascii Either a matrix (by columns) or CSV file or connection (by rows) +#' @param data_bin_file Name of binary file on output of (\code{binarize}) +#' or input of (\code{getDataInFile}) #' @param nb_per_chunk Number of lines to process in one batch (big.matrix or connection) -#' @inheritParams claws #' @param getData Function to retrieve data chunks #' @param transform Transformation function to apply on data chunks +#' @param indices Indices of the lines to retrieve +#' @inheritParams claws #' #' @return For \code{getDataInFile()}, the matrix with rows corresponding to the #' requested indices. \code{binarizeTransform} returns the number of processed lines. #' \code{binarize} is designed to serialize in several calls, thus returns nothing. +#' +#' @name de_serialize +#' @rdname de_serialize +#' @aliases binarize binarizeTransform getDataInFile NULL #' @rdname de_serialize