Wrapper for dplyr's summarize

This function wraps dplyr's summarize() function in a convenient way. The user only needs to define functions on the dataset with a named vector or list (with atomic entries of length 1) as return.

dplyr_wrapper(data, group_by, fun, check_fun = TRUE)

Arguments

data	(`dataframe`). A dataframe with a grouping variable.
group_by	(`character()`). Name of column, which contains identifiers on which the dataset should be grouped by. E.g. different user IDs.
fun	(`function`). Must be a function, which has a dataframe as input and a (named) vector of desired length as output.
check_fun	(`logical(1)`). If TRUE, fun(data) will be evaluated and checked if the outcome is of correct form. Set to FALSE if evaluation on the whole dataset takes too long.

Value

(`dataframe`)

Examples

# Number of used chrome apps
fun1 = function(data) {
  c(uses_chrome = nrow(
    dplyr::filter(data, RUNNING_TASKS_baseActivity_mPackage == "com.android.chrome"))
  )
}
dplyr_wrapper(data = studentlife_small, group_by = "userId", fun = fun1)
#>   userId uses_chrome
#> 1     00        3104
#> 2     01        1635
#> 3     02        2811

# mean, max, sd of a column
fun2 = function(data) {
  c(mean_sepal_length = mean(data$Sepal.Length),
    max_sepal_length = max(data$Sepal.Length),
    sd_sepal_length = sd(data$Sepal.Length)
  )
}
dplyr_wrapper(data = iris, group_by = "Species", fun = fun2)
#>      Species mean_sepal_length max_sepal_length sd_sepal_length
#> 1     setosa             5.006              5.8       0.3524897
#> 2 versicolor             5.936              7.0       0.5161711
#> 3  virginica             6.588              7.9       0.6358796

# return list
fun3 = function(data) {
  list(mean_sepal_length = mean(data$Sepal.Length),
    max_sepal_length = max(data$Sepal.Length),
    sd_sepal_length = sd(data$Sepal.Length)
  )
}
dplyr_wrapper(data = iris, group_by = "Species", fun = fun3)
#>      Species mean_sepal_length max_sepal_length sd_sepal_length
#> 1     setosa             5.006              5.8       0.3524897
#> 2 versicolor             5.936              7.0       0.5161711
#> 3  virginica             6.588              7.9       0.6358796

# group by two columns
df = data.frame(id = c(rep(1, 10), rep(2, 10)))
df$task = rep(c(rep("task1", 5), rep("task2", 5)), 2)
df$hour = rep(c(rep("hour1", 3), rep("hour2", 2), rep("hour1", 2), rep("hour2", 3)), 2)
df$x = 1:20
fun4 = function(data) c(mean_x = mean(data$x))
dplyr_wrapper(data = df, group_by = c("id", "task"), fun = fun4)
#> Warning: `group_by_()` is deprecated as of dplyr 0.7.0.
#> Please use `group_by()` instead.
#> See vignette('programming') for more help
#> This warning is displayed once every 8 hours.
#> Call `lifecycle::last_warnings()` to see where this warning was generated.
#>   id task1 task2
#> 1  1     3     8
#> 2  2    13    18