Github user junyangq commented on a diff in the pull request: https://github.com/apache/spark/pull/14384#discussion_r74112930 --- Diff: R/pkg/R/mllib.R --- @@ -632,3 +642,147 @@ setMethod("predict", signature(object = "AFTSurvivalRegressionModel"), function(object, newData) { return(dataFrame(callJMethod(object@jobj, "transform", newData@sdf))) }) + + +#' Alternating Least Squares (ALS) for Collaborative Filtering +#' +#' \code{spark.als} learns latent factors in collaborative filtering via alternating least +#' squares. Users can call \code{summary} to obtain fitted latent factors, \code{predict} +#' to make predictions on new data, and \code{write.ml}/\code{read.ml} to save/load fitted models. +#' +#' For more details, see +#' \href{http://spark.apache.org/docs/latest/ml-collaborative-filtering.html}{MLlib: +#' Collaborative Filtering}. +#' Additional arguments can be passed to the methods. +#' \describe{ +#' \item{nonnegative}{logical value indicating whether to apply nonnegativity constraints. +#' Default: FALSE} +#' \item{implicitPrefs}{logical value indicating whether to use implicit preference. +#' Default: FALSE} +#' \item{alpha}{alpha parameter in the implicit preference formulation (>= 0). Default: 1.0} +#' \item{seed}{integer seed for random number generation. Default: 0} +#' \item{numUserBlocks}{number of user blocks used to parallelize computation (> 0). +#' Default: 10} +#' \item{numItemBlocks}{number of item blocks used to parallelize computation (> 0). +#' Default: 10} +#' \item{checkpointInterval}{number of checkpoint intervals (>= 1) or disable checkpoint (-1). +#' Default: 10} +#' } +#' +#' @param data A SparkDataFrame for training +#' @param ratingCol column name for ratings +#' @param userCol column name for user ids. Ids must be (or can be coerced into) integers +#' @param itemCol column name for item ids. Ids must be (or can be coerced into) integers +#' @param rank rank of the matrix factorization (> 0) +#' @param reg regularization parameter (>= 0) +#' @param maxIter maximum number of iterations (>= 0) + +#' @return \code{spark.als} returns a fitted ALS model +#' @rdname spark.als +#' @aliases spark.als,SparkDataFrame +#' @name spark.als +#' @export +#' @examples +#' \dontrun{ +#' df <- createDataFrame(ratings) +#' model <- spark.als(df, "rating", "user", "item") +#' +#' # extract latent factors +#' stats <- summary(model) +#' userFactors <- stats$userFactors +#' itemFactors <- stats$itemFactors +#' +#' # make predictions +#' predicted <- predict(model, df) +#' showDF(predicted) +#' +#' # save and load the model +#' path <- "path/to/model" +#' write.ml(model, path) +#' savedModel <- read.ml(path) +#' summary(savedModel) +#' +#' # set other arguments +#' modelS <- spark.als(df, "rating", "user", "item", rank = 20, +#' reg = 0.1, nonnegative = TRUE) +#' statsS <- summary(modelS) +#' } +#' @note spark.als since 2.1.0 +setMethod("spark.als", signature(data = "SparkDataFrame"), + function(data, ratingCol = "rating", userCol = "user", itemCol = "item", + rank = 10, reg = 1.0, maxIter = 10, ...) { + + `%||%` <- function(a, b) if (!is.null(a)) a else b --- End diff -- In this case (since we set 7 default values) the code would be a little repetitive if we expand every one of them. I guess this definition would not have side effect to the other functions in SparkR?
--- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your project does not have this feature enabled and wishes so, or if the feature is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. --- --------------------------------------------------------------------- To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org