gapply {SparkR}R Documentation

gapply

Description

Groups the SparkDataFrame using the specified columns and applies the R function to each group.

Usage

gapply(x, ...)

## S4 method for signature 'GroupedData'
gapply(x, func, schema)

## S4 method for signature 'SparkDataFrame'
gapply(x, cols, func, schema)

Arguments

x

a SparkDataFrame or GroupedData.

...

additional argument(s) passed to the method.

func

a function to be applied to each group partition specified by grouping column of the SparkDataFrame. See Details.

schema

the schema of the resulting SparkDataFrame after the function is applied. The schema must match to output of func. It has to be defined for each output column with preferred output column name and corresponding data type. Since Spark 2.3, the DDL-formatted string is also supported for the schema.

cols

grouping columns.

Details

func is a function of two arguments. The first, usually named key (though this is not enforced) corresponds to the grouping key, will be an unnamed list of length(cols) length-one objects corresponding to the grouping columns' values for the current group.

The second, herein x, will be a local data.frame with the columns of the input not in cols for the rows corresponding to key.

The output of func must be a data.frame matching schema – in particular this means the names of the output data.frame are irrelevant

Value

A SparkDataFrame.

Note

gapply(GroupedData) since 2.0.0

gapply(SparkDataFrame) since 2.0.0

See Also

gapplyCollect

Other SparkDataFrame functions: SparkDataFrame-class, agg(), alias(), arrange(), as.data.frame(), attach,SparkDataFrame-method, broadcast(), cache(), checkpoint(), coalesce(), collect(), colnames(), coltypes(), createOrReplaceTempView(), crossJoin(), cube(), dapplyCollect(), dapply(), describe(), dim(), distinct(), dropDuplicates(), dropna(), drop(), dtypes(), exceptAll(), except(), explain(), filter(), first(), gapplyCollect(), getNumPartitions(), group_by(), head(), hint(), histogram(), insertInto(), intersectAll(), intersect(), isLocal(), isStreaming(), join(), limit(), localCheckpoint(), merge(), mutate(), ncol(), nrow(), persist(), printSchema(), randomSplit(), rbind(), rename(), repartitionByRange(), repartition(), rollup(), sample(), saveAsTable(), schema(), selectExpr(), select(), showDF(), show(), storageLevel(), str(), subset(), summary(), take(), toJSON(), unionAll(), unionByName(), union(), unpersist(), withColumn(), withWatermark(), with(), write.df(), write.jdbc(), write.json(), write.orc(), write.parquet(), write.stream(), write.text()

Examples

## Not run: 
##D # Computes the arithmetic mean of the second column by grouping
##D # on the first and third columns. Output the grouping values and the average.
##D 
##D df <- createDataFrame (
##D list(list(1L, 1, "1", 0.1), list(1L, 2, "1", 0.2), list(3L, 3, "3", 0.3)),
##D   c("a", "b", "c", "d"))
##D 
##D # Here our output contains three columns, the key which is a combination of two
##D # columns with data types integer and string and the mean which is a double.
##D schema <- structType(structField("a", "integer"), structField("c", "string"),
##D   structField("avg", "double"))
##D result <- gapply(
##D   df,
##D   c("a", "c"),
##D   function(key, x) {
##D     # key will either be list(1L, '1') (for the group where a=1L,c='1') or
##D     #   list(3L, '3') (for the group where a=3L,c='3')
##D     y <- data.frame(key, mean(x$b), stringsAsFactors = FALSE)
##D }, schema)
##D 
##D # The schema also can be specified in a DDL-formatted string.
##D schema <- "a INT, c STRING, avg DOUBLE"
##D result <- gapply(
##D   df,
##D   c("a", "c"),
##D   function(key, x) {
##D     y <- data.frame(key, mean(x$b), stringsAsFactors = FALSE)
##D }, schema)
##D 
##D # We can also group the data and afterwards call gapply on GroupedData.
##D # For example:
##D gdf <- group_by(df, "a", "c")
##D result <- gapply(
##D   gdf,
##D   function(key, x) {
##D     y <- data.frame(key, mean(x$b), stringsAsFactors = FALSE)
##D }, schema)
##D collect(result)
##D 
##D # Result
##D # ------
##D # a c avg
##D # 3 3 3.0
##D # 1 1 1.5
##D 
##D # Fits linear models on iris dataset by grouping on the 'Species' column and
##D # using 'Sepal_Length' as a target variable, 'Sepal_Width', 'Petal_Length'
##D # and 'Petal_Width' as training features.
##D 
##D df <- createDataFrame (iris)
##D schema <- structType(structField("(Intercept)", "double"),
##D   structField("Sepal_Width", "double"),structField("Petal_Length", "double"),
##D   structField("Petal_Width", "double"))
##D df1 <- gapply(
##D   df,
##D   df$"Species",
##D   function(key, x) {
##D     m <- suppressWarnings(lm(Sepal_Length ~
##D     Sepal_Width + Petal_Length + Petal_Width, x))
##D     data.frame(t(coef(m)))
##D   }, schema)
##D collect(df1)
##D 
##D # Result
##D # ---------
##D # Model  (Intercept)  Sepal_Width  Petal_Length  Petal_Width
##D # 1        0.699883    0.3303370    0.9455356    -0.1697527
##D # 2        1.895540    0.3868576    0.9083370    -0.6792238
##D # 3        2.351890    0.6548350    0.2375602     0.2521257
##D 
## End(Not run)

[Package SparkR version 3.0.0 Index]