Skip to contents

Aggregate functions defined for Column.

Usage

approx_count_distinct(x, ...)

approxCountDistinct(x, ...)

collect_list(x)

collect_set(x)

count_distinct(x, ...)

countDistinct(x, ...)

grouping_bit(x)

grouping_id(x, ...)

kurtosis(x)

max_by(x, y)

min_by(x, y)

n_distinct(x, ...)

percentile_approx(x, percentage, ...)

product(x)

sd(x, na.rm = FALSE)

skewness(x)

stddev(x)

std(x)

stddev_pop(x)

stddev_samp(x)

sum_distinct(x)

sumDistinct(x)

var(x, y = NULL, na.rm = FALSE, use)

variance(x)

var_pop(x)

var_samp(x)

# S4 method for class 'Column'
approx_count_distinct(x, rsd = 0.05)

# S4 method for class 'Column'
approxCountDistinct(x, rsd = 0.05)

# S4 method for class 'Column'
kurtosis(x)

# S4 method for class 'Column'
max(x)

# S4 method for class 'Column,Column'
max_by(x, y)

# S4 method for class 'Column'
mean(x)

# S4 method for class 'Column'
min(x)

# S4 method for class 'Column,Column'
min_by(x, y)

# S4 method for class 'Column'
product(x)

# S4 method for class 'characterOrColumn,numericOrColumn'
percentile_approx(x, percentage, accuracy = 10000)

# S4 method for class 'Column'
sd(x)

# S4 method for class 'Column'
skewness(x)

# S4 method for class 'Column'
stddev(x)

# S4 method for class 'Column'
std(x)

# S4 method for class 'Column'
stddev_pop(x)

# S4 method for class 'Column'
stddev_samp(x)

# S4 method for class 'Column'
sum(x)

# S4 method for class 'Column'
sum_distinct(x)

# S4 method for class 'Column'
sumDistinct(x)

# S4 method for class 'Column'
var(x)

# S4 method for class 'Column'
variance(x)

# S4 method for class 'Column'
var_pop(x)

# S4 method for class 'Column'
var_samp(x)

# S4 method for class 'Column'
approx_count_distinct(x, rsd = 0.05)

# S4 method for class 'Column'
approxCountDistinct(x, rsd = 0.05)

# S4 method for class 'Column'
count_distinct(x, ...)

# S4 method for class 'Column'
countDistinct(x, ...)

# S4 method for class 'Column'
n_distinct(x, ...)

# S4 method for class 'Column'
collect_list(x)

# S4 method for class 'Column'
collect_set(x)

# S4 method for class 'Column'
grouping_bit(x)

# S4 method for class 'Column'
grouping_id(x, ...)

Arguments

x

Column to compute on.

...

additional argument(s). For example, it could be used to pass additional Columns.

y, na.rm, use

currently not used.

percentage

Numeric percentage at which percentile should be computed All values should be between 0 and 1. If length equals to 1 resulting column is of type double, otherwise, array type of double.

rsd

maximum relative standard deviation allowed (default = 0.05).

accuracy

A positive numeric literal (default: 10000) which controls approximation accuracy at the cost of memory. Higher value of accuracy yields better accuracy, 1.0/accuracy is the relative error of the approximation.

Details

approx_count_distinct: Returns the approximate number of distinct items in a group.

approxCountDistinct: Returns the approximate number of distinct items in a group.

kurtosis: Returns the kurtosis of the values in a group.

max: Returns the maximum value of the expression in a group.

max_by: Returns the value associated with the maximum value of ord.

Note: The function is non-deterministic so the output order can be different for those associated the same values of `x`.

mean: Returns the average of the values in a group. Alias for avg.

min: Returns the minimum value of the expression in a group.

min_by: Returns the value associated with the minimum value of ord.

Note: The function is non-deterministic so the output order can be different for those associated the same values of `x`.

product: Returns the product of the values in a group.

percentile_approx Returns the approximate percentile of the numeric column col which is the smallest value in the ordered col values (sorted from least to greatest) such that no more than percentage of col values is less than the value or equal to that value.

sd: Alias for stddev_samp.

skewness: Returns the skewness of the values in a group.

stddev: Alias for std_dev.

std: Alias for stddev.

stddev_pop: Returns the population standard deviation of the expression in a group.

stddev_samp: Returns the unbiased sample standard deviation of the expression in a group.

sum: Returns the sum of all values in the expression.

sum_distinct: Returns the sum of distinct values in the expression.

sumDistinct: Returns the sum of distinct values in the expression.

var: Alias for var_samp.

var_pop: Returns the population variance of the values in a group.

var_samp: Returns the unbiased variance of the values in a group.

count_distinct: Returns the number of distinct items in a group.

countDistinct: Returns the number of distinct items in a group.

An alias of count_distinct, and it is encouraged to use count_distinct directly.

n_distinct: Returns the number of distinct items in a group.

collect_list: Creates a list of objects with duplicates. Note: the function is non-deterministic because the order of collected results depends on the order of the rows which may be non-deterministic after a shuffle.

collect_set: Creates a list of objects with duplicate elements eliminated. Note: the function is non-deterministic because the order of collected results depends on the order of the rows which may be non-deterministic after a shuffle.

grouping_bit: Indicates whether a specified column in a GROUP BY list is aggregated or not, returns 1 for aggregated or 0 for not aggregated in the result set. Same as GROUPING in SQL and grouping function in Scala.

grouping_id: Returns the level of grouping. Equals to grouping_bit(c1) * 2^(n - 1) + grouping_bit(c2) * 2^(n - 2) + ... + grouping_bit(cn) .

Note

approx_count_distinct(Column) since 3.0.0

approxCountDistinct(Column) since 1.4.0

kurtosis since 1.6.0

max since 1.5.0

max_by since 3.3.0

mean since 1.5.0

min since 1.5.0

min_by since 3.3.0

product since 3.2.0

percentile_approx since 3.1.0

sd since 1.6.0

skewness since 1.6.0

stddev since 1.6.0

std since 3.5.0

stddev_pop since 1.6.0

stddev_samp since 1.6.0

sum since 1.5.0

sum_distinct since 3.2.0

sumDistinct since 1.4.0

var since 1.6.0

variance since 1.6.0

var_pop since 1.5.0

var_samp since 1.6.0

approx_count_distinct(Column, numeric) since 3.0.0

approxCountDistinct(Column, numeric) since 1.4.0

count_distinct since 3.2.0

countDistinct since 1.4.0

n_distinct since 1.4.0

collect_list since 2.3.0

collect_set since 2.3.0

grouping_bit since 2.3.0

grouping_id since 2.3.0

See also

Other aggregate functions: avg(), corr(), count(), cov(), first(), last()

Examples

if (FALSE) { # \dontrun{
# Dataframe used throughout this doc
df <- createDataFrame(cbind(model = rownames(mtcars), mtcars))} # }

if (FALSE) { # \dontrun{
head(select(df, approx_count_distinct(df$gear)))
head(select(df, approx_count_distinct(df$gear, 0.02)))
head(select(df, count_distinct(df$gear, df$cyl)))
head(select(df, n_distinct(df$gear)))
head(distinct(select(df, "gear")))} # }

if (FALSE) { # \dontrun{
head(select(df, mean(df$mpg), sd(df$mpg), skewness(df$mpg), kurtosis(df$mpg)))} # }

if (FALSE) { # \dontrun{
df <- createDataFrame(
  list(list("Java", 2012, 20000), list("dotNET", 2012, 5000),
       list("dotNET", 2013, 48000), list("Java", 2013, 30000)),
  list("course", "year", "earnings")
)
tmp <- agg(groupBy(df, df$"course"), "max_by" = max_by(df$"year", df$"earnings"))
head(tmp)} # }

if (FALSE) { # \dontrun{
head(select(df, avg(df$mpg), mean(df$mpg), sum(df$mpg), min(df$wt), max(df$qsec)))

# metrics by num of cylinders
tmp <- agg(groupBy(df, "cyl"), avg(df$mpg), avg(df$hp), avg(df$wt), avg(df$qsec))
head(orderBy(tmp, "cyl"))

# car with the max mpg
mpg_max <- as.numeric(collect(agg(df, max(df$mpg))))
head(where(df, df$mpg == mpg_max))} # }

if (FALSE) { # \dontrun{
df <- createDataFrame(
  list(list("Java", 2012, 20000), list("dotNET", 2012, 5000),
       list("dotNET", 2013, 48000), list("Java", 2013, 30000)),
  list("course", "year", "earnings")
)
tmp <- agg(groupBy(df, df$"course"), "min_by" = min_by(df$"year", df$"earnings"))
head(tmp)} # }

if (FALSE) { # \dontrun{
head(select(df, sd(df$mpg), stddev(df$mpg), stddev_pop(df$wt), stddev_samp(df$qsec)))} # }

if (FALSE) { # \dontrun{
head(select(df, sum_distinct(df$gear)))
head(distinct(select(df, "gear")))} # }

if (FALSE) { # \dontrun{
head(agg(df, var(df$mpg), variance(df$mpg), var_pop(df$mpg), var_samp(df$mpg)))} # }

if (FALSE) { # \dontrun{
df2 = df[df$mpg > 20, ]
collect(select(df2, collect_list(df2$gear)))
collect(select(df2, collect_set(df2$gear)))} # }

if (FALSE) { # \dontrun{
# With cube
agg(
  cube(df, "cyl", "gear", "am"),
  mean(df$mpg),
  grouping_bit(df$cyl), grouping_bit(df$gear), grouping_bit(df$am)
)

# With rollup
agg(
  rollup(df, "cyl", "gear", "am"),
  mean(df$mpg),
  grouping_bit(df$cyl), grouping_bit(df$gear), grouping_bit(df$am)
)} # }

if (FALSE) { # \dontrun{
# With cube
agg(
  cube(df, "cyl", "gear", "am"),
  mean(df$mpg),
  grouping_id(df$cyl, df$gear, df$am)
)

# With rollup
agg(
  rollup(df, "cyl", "gear", "am"),
  mean(df$mpg),
  grouping_id(df$cyl, df$gear, df$am)
)} # }