Merge pull request #245 from hqzizania/upstream

Add Rd files for sampleByKey() of [SPARKR-163] and sumRDD() of [SPARKR-92]

Author: shivaram

R/pkg/R/pairRDD.R

@@ -821,8 +821,11 @@ setMethod("subtractByKey",
                           function (v) { v[[1]] })
           })
 
+#' Return a subset of this RDD sampled by key.
+#' 
 #' @description
-#' \code{sampleByKey} return a subset RDD of the given RDD sampled by key
+#' \code{sampleByKey} Create a sample of this RDD using variable sampling rates
+#' for different keys as specified by fractions, a key to sampling rate map.
 #'
 #' @param x The RDD to sample elements by key, where each element is
 #'             list(K, V) or c(K, V).

pkg/man/sampleByKey.Rd

@@ -0,0 +1,52 @@
+% Generated by roxygen2 (4.1.0): do not edit by hand
+% Please edit documentation in R/generics.R, R/pairRDD.R
+\docType{methods}
+\name{sampleByKey}
+\alias{sampleByKey}
+\alias{sampleByKey,RDD,logical,vector,integer-method}
+\alias{sampleByKey,RDD-method}
+\title{Return a subset of this RDD sampled by key.}
+\usage{
+sampleByKey(x, withReplacement, fractions, seed)
+
+\S4method{sampleByKey}{RDD,logical,vector,integer}(x, withReplacement,
+  fractions, seed)
+}
+\arguments{
+\item{x}{The RDD to sample elements by key, where each element is
+list(K, V) or c(K, V).}
+
+\item{withReplacement}{Sampling with replacement or not}
+
+\item{seed}{Randomness seed value}
+
+\item{fraction}{The (rough) sample target fraction}
+}
+\description{
+\code{sampleByKey} Create a sample of this RDD using variable sampling rates
+for different keys as specified by fractions, a key to sampling rate map.
+}
+\examples{
+\dontrun{
+sc <- sparkR.init()
+rdd <- parallelize(sc, 1:3000)
+pairs <- lapply(rdd, function(x) { if (x \%\% 3 == 0) list("a", x)
+                                   else { if (x \%\% 3 == 1) list("b", x) else list("c", x) }})
+fractions <- list(a = 0.2, b = 0.1, c = 0.3)
+sample <- sampleByKey(pairs, FALSE, fractions, 1618L)
+100 < length(lookup(sample, "a")) && 300 > length(lookup(sample, "a")) # TRUE
+50 < length(lookup(sample, "b")) && 150 > length(lookup(sample, "b")) # TRUE
+200 < length(lookup(sample, "c")) && 400 > length(lookup(sample, "c")) # TRUE
+lookup(sample, "a")[which.min(lookup(sample, "a"))] >= 0 # TRUE
+lookup(sample, "a")[which.max(lookup(sample, "a"))] <= 2000 # TRUE
+lookup(sample, "b")[which.min(lookup(sample, "b"))] >= 0 # TRUE
+lookup(sample, "b")[which.max(lookup(sample, "b"))] <= 2000 # TRUE
+lookup(sample, "c")[which.min(lookup(sample, "c"))] >= 0 # TRUE
+lookup(sample, "c")[which.max(lookup(sample, "c"))] <= 2000 # TRUE
+fractions <- list(a = 0.2, b = 0.1, c = 0.3, d = 0.4)
+sample <- sampleByKey(pairs, FALSE, fractions, 1618L) # Key "d" will be ignored
+fractions <- list(a = 0.2, b = 0.1)
+sample <- sampleByKey(pairs, FALSE, fractions, 1618L) # KeyError: "c"
+}
+}
+

pkg/man/sumRDD.Rd

@@ -0,0 +1,27 @@
+% Generated by roxygen2 (4.1.0): do not edit by hand
+% Please edit documentation in R/RDD.R, R/generics.R
+\docType{methods}
+\name{sumRDD,RDD-method}
+\alias{sumRDD}
+\alias{sumRDD,RDD}
+\alias{sumRDD,RDD-method}
+\title{Add up the elements in an RDD.}
+\usage{
+\S4method{sumRDD}{RDD}(x)
+
+sumRDD(x)
+}
+\arguments{
+\item{x}{The RDD to add up the elements in}
+}
+\description{
+Add up the elements in an RDD.
+}
+\examples{
+\dontrun{
+sc <- sparkR.init()
+rdd <- parallelize(sc, 1:10)
+sumRDD(rdd) # 55
+}
+}
+