Using tidyselect in your own package
The {tidyselect} package powers column selection for {dplyr}, {tidyr} and {recipes} functions but it's also quite straight-forward to include its functonality in other user-built functions. Here's an example which replaces a percentage of a data frame by NA. It works by using the tidyselect::eval_select function to select columns and then replaces a portion of the data with NA values.
Column selections are passed to the function through dot-dot-dot, allowing an arbitrary number of columns to be selected. The first line of the function:
1expr <- rlang::expr(c(...))
returns a defused expression, and the second:
1pos <- tidyselect::eval_select(expr, data = data)
resumes execution, returning a vector of positions that match the selection. More details on how to implement tidyselect interfaces is available at the tidyselect pkgdown page.
1#' populate data frame with missing data
2#'
3#' Replace a proportion of data in a data frame with missing values
4#'
5#' @param data A data frame
6#' @param p proporion (between 0 and 100) of data in column to be flagged as missing
7#' @param ... `<tidy-select>` One or more unquoted expressions separated by commas
8#'
9#' @return data frame
10#'
11#' @importFrom rlang expr
12#' @importFrom tidyselect vars_select
13#' @export
14add_missing_df <- function(data, ..., p = 10) {
15 expr <- rlang::expr(c(...))
16 pos <- tidyselect::eval_select(expr, data = data)
17 if (length(pos) > 0) {
18 for (posn in pos) {
19 missing_rows <- sample(nrow(data), size = as.integer(nrow(data) * p / 100))
20 if (length(missing_rows) > 0) {
21 data[missing_rows, posn] <- NA
22 }
23 }
24 }
25 return(data)
26}
Examples
All of the {tidyselect} selectors may be used, along with the {magrittr} pipe if loaded. Some examples of use are:
1# remove 40% of data in columns mpg to hp
2df <- add_missing_df(mtcars, p=40, mpg:hp)
3head(df)
| mpg | cyl | disp | hp | drat | wt | qsec | vs | am | gear | carb | |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Mazda RX4 | 21.0 | 6 | NA | NA | 3.90 | 2.620 | 16.46 | 0 | 1 | 4 | 4 |
| Mazda RX4 Wag | 21.0 | NA | 160 | NA | 3.90 | 2.875 | 17.02 | 0 | 1 | 4 | 4 |
| Datsun 710 | 22.8 | NA | NA | NA | 3.85 | 2.320 | 18.61 | 1 | 1 | 4 | 1 |
| Hornet 4 Drive | NA | NA | 258 | 110 | 3.08 | 3.215 | 19.44 | 1 | 0 | 3 | 1 |
| Hornet Sportabout | NA | 8 | NA | 175 | 3.15 | 3.440 | 17.02 | 0 | 0 | 3 | 2 |
| Valiant | 18.1 | NA | 225 | 105 | 2.76 | 3.460 | 20.22 | 1 | 0 | 3 | 1 |
1# remove 10% of data in columns that start with "d" (dist and drat)
2df <- mtcars %>% add_missing_df(startswith("d"))
3head(df)
| mpg | cyl | disp | hp | drat | wt | qsec | vs | am | gear | carb | |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Mazda RX4 | 21.0 | 6 | NA | 110 | NA | 2.620 | 16.46 | 0 | 1 | 4 | 4 |
| Mazda RX4 Wag | 21.0 | 6 | 160 | 110 | 3.90 | 2.875 | 17.02 | 0 | 1 | 4 | 4 |
| Datsun 710 | 22.8 | 4 | 108 | 93 | 3.85 | 2.320 | 18.61 | 1 | 1 | 4 | 1 |
| Hornet 4 Drive | 21.4 | 6 | 258 | 110 | 3.08 | 3.215 | 19.44 | 1 | 0 | 3 | 1 |
| Hornet Sportabout | 18.7 | 8 | 360 | 175 | NA | 3.440 | 17.02 | 0 | 0 | 3 | 2 |
| Valiant | 18.1 | 6 | 225 | 105 | 2.76 | 3.460 | 20.22 | 1 | 0 | 3 | 1 |