Practical 1

Introducing purrr and map

1. Write a function that calculates the square of a number (i.e., \(x^2\)).
2. Create a vector x containing the numbers 1 to 10.
3. Use map() to apply your function to each number in the vector.
4. Repeat, but this time use a temporary function.

NoteSolution

1library(tidyverse)
2sq <- function(x) { x^2 }
3x <- 1:10

4map(x, sq)

5map(x, \(x) x^2)

1

We need to load the tidyverse package (or purrr).

2

Write a function to calculate the square.

3

Define a vector of numbers.

4

Use map to apply the function to the vector.

5

Use a temporary function (\(x) x^2) to calculate the square of each number.

5. Modify your previous answer so that the result is returned as a numeric vector instead of a list.

NoteSolution

map_dbl(x, \(x) x^2)

 [1]   1   4   9  16  25  36  49  64  81 100

6. Write a vector of names (e.g., c("Alice", "Bob")).
7. Use map_chr() to convert this vector into greetings (e.g., “Hello, Alice!”).

NoteSolution

people <- c("Alice", "Bob", "Charlie")
map_chr(people, \(x) paste0("Hello ", x, "!"))

[1] "Hello Alice!"   "Hello Bob!"     "Hello Charlie!"

# We can also use `str_glue` for string interpolation:
map_chr(people, \(x) str_glue("Hello, {x}!"))

[1] "Hello, Alice!"   "Hello, Bob!"     "Hello, Charlie!"

8. Given the below list of numeric vectors, use map_dbl() to compute the mean of each vector.

numbers <- list(c(1, 2, 3), c(4, 5, 6), c(7, 8, 9))

NoteSolution

map_dbl(numbers, mean)

[1] 2 5 8

9. Use the code below to generate 10 datasets and save them into a subfolder of your current working directory.

generate_data <- function() {
1  d <- as.data.frame(matrix(rnorm(500), ncol = 5))
2  names(d) <- paste0("var", 1:ncol(d))
  return(d)
}

3dir.create("datasets")

4walk(LETTERS[1:10], \(i) {
5  write_csv(generate_data(),
            file = paste0("datasets/", i, ".csv")) 
})

1

Create a matrix from a vector of 500 random numbers and convert it to a data frame.

2

Set the names of the data frame to var1, var2, etc.

3

Create a new folder “datasets”

4

For each letter from A to J

5

Generate a dataset and save it as a CSV file

---

10. Use map to import all CSV files into this folder as a list of data frames.

11. Use list_rbind to convert the list into a single data frame. Use the names_to argument to retain the dataset names.

NoteSolution

1library(fs)

2result <- dir_ls("datasets") |>
3  map(read_csv) |>
4  list_rbind(names_to = "file")

# To extract the dataset name (i.e., A, B, C, ...)
result |>
  mutate(file = str_sub(file, 10, 10))

1

fs is a powerful, flexible package for working with the file system.

2

Use dir_ls to get a list of files in the ‘datasets’ folder.

3

Apply read_csv to each filename.

4

Combine the list of data frames by row-binding.