4 Data frames and tibbles
4.2 Reshaping wide and long data
Task: Gather/melt/unpivot/fold from wide into long format (two columns).
data.frame(abc = sample(letters[1:3], 6, replace = TRUE),
r1 = rnorm(6),
r2 = rnorm(6),
stringsAsFactors = FALSE) %>%
tidyr::gather(key = r, value = random, r1, r2) %>%
head(4)
#> abc r random
#> 1 c r1 -0.1607019
#> 2 a r1 0.1618228
#> 3 a r1 1.1763865
#> 4 b r1 -0.9474683
# ____________________________________________________________________________Task: Spread/cast/pivot/unfold from long into wide format (two columns).
data.frame(abc = rep(letters[1:4], times = 2),
r = rep(c("r1", "r2"), each = 4),
random = rnorm(8),
stringsAsFactors = FALSE) %>%
tidyr::spread(key = abc, value = random) %>%
head(1)
#> r a b c d
#> 1 r1 0.5033555 0.05401 -1.484874 -0.1451563
# ____________________________________________________________________________4.3 Typical mutate statements
4.3.1 Split and paste
Task: Split/separate a column into one or more new columns.
iris %>% dplyr::transmute(Sepal.Length = as.character(Sepal.Length)) %>%
tidyr::separate(col = Sepal.Length,
into = c("Sep1", "Sep2"),
sep = "\\.",
remove = FALSE) %>%
head(1)
#> Warning: Too few values at 17 locations: 5, 8, 26, 27, 36, 41, 44, 50, 51,
#> 61, 63, 79, 84, 86, 94, 120, 139
#> Sepal.Length Sep1 Sep2
#> 1 5.1 5 1Task: Combine/unite/paste one or more columns into one.
iris[1,] %>% tidyr::unite(new_col, Sepal.Width, Species, sep = "", remove = FALSE)
#> Sepal.Length new_col Sepal.Width Petal.Length Petal.Width Species
#> 1 5.1 3.5setosa 3.5 1.4 0.2 setosa
iris[1,] %>% dplyr::mutate(new_col = stringr::str_c(Sepal.Width, Species))
#> Sepal.Length Sepal.Width Petal.Length Petal.Width Species new_col
#> 1 5.1 3.5 1.4 0.2 setosa 3.5setosa
# ____________________________________________________________________________4.3.2 Conditionals
Task: Mutate columns depending/conditionally on other colums.
# use dplyr::case_when() or dplyr::if_else()
# ____________________________________________________________________________
# use ifelse4.3.3 IDs
Task Add an id column.
iris[1:2, ] %>%
dplyr::mutate(id = seq_len(n()))
#> Sepal.Length Sepal.Width Petal.Length Petal.Width Species id
#> 1 5.1 3.5 1.4 0.2 setosa 1
#> 2 4.9 3.0 1.4 0.2 setosa 2
# ____________________________________________________________________________
df <- data.frame(a = 1:2)
df[["id"]] <- seq_len(nrow(df))Task Add a unique identifier regarding some columns (not in order).
iris %>%
dplyr::mutate(id = as.integer(factor(stringr::str_c(Species, Sepal.Width))))
# ____________________________________________________________________________
iris[["id"]] <- as.integer(factor(paste0(iris[["Species"]], iris[["Sepal.Width"]])))Task Add a unique identifier regarding all (or some) columns, in order off appearing unique rows.
iris %>% dplyr::mutate(id = iris %>%
purrr::pmap_chr(stringr::str_c, sep = "\t") %>%
factor %>%
forcats::fct_inorder(ordered = TRUE) %>%
as.integer)
# ____________________________________________________________________________4.3.4 Other stuff
Task: Use some helpers to select columns based on naming patterns
iris[1,] %>%
dplyr::select(dplyr::starts_with("S",
ignore.case = TRUE,
vars = dplyr::current_vars()))
#> Sepal.Length Sepal.Width Species
#> 1 5.1 3.5 setosa
iris[1,] %>% dplyr::select(dplyr::ends_with("h"))
#> Sepal.Length Sepal.Width Petal.Length Petal.Width
#> 1 5.1 3.5 1.4 0.2
iris[1,] %>% dplyr::select(dplyr::contains("."))
#> Sepal.Length Sepal.Width Petal.Length Petal.Width
#> 1 5.1 3.5 1.4 0.2
iris[1,] %>% dplyr::select(dplyr::matches("."))
#> Sepal.Length Sepal.Width Petal.Length Petal.Width Species
#> 1 5.1 3.5 1.4 0.2 setosa
purrr::set_names(iris[1,], paste0("a", 1:5)) %>%
dplyr::select(dplyr::num_range(prefix = "a", range = 2:4, width = 1))
#> a2 a3 a4
#> 1 3.5 1.4 0.2
iris[1, ] %>% dplyr::select(dplyr::one_of("Sepal.Width", "Species"))
#> Sepal.Width Species
#> 1 3.5 setosa
# ____________________________________________________________________________Task: Summarise grouped data by a statistic that returns more than one value.
Task: Add summary data directly to the summarised data frame.
4.4 Formatting
Task: Change column names to lower case.
stringr::str_to_lower(names(iris))
#> [1] "sepal.length" "sepal.width" "petal.length" "petal.width"
#> [5] "species"
# ____________________________________________________________________________
tolower(names(iris))
#> [1] "sepal.length" "sepal.width" "petal.length" "petal.width"
#> [5] "species"Task: Change column names to snake_case.
library(magrittr)
c("Var 1", "Var-2", "Var.3", "Var4") %>%
stringr::str_replace_all("\\s|-|\\.", "_") %>%
stringr::str_to_lower()
#> [1] "var_1" "var_2" "var_3" "var4"Task: Change column names from camelCase to snake_case.
to_snake_case <- function(camelcases){
# catch some input that should be handled like underscores too
camelcases <- stringr::str_replace_all(camelcases, "\\s+|\\.+", "_")
# get to know, if a string starts with a small letter
small_start <- !is.na(stringr::str_extract(camelcases, "^[a-z]"))
# get all capital letter sequences from a string
capitals <- stringr::str_extract_all(camelcases, "[A-Z]+")
# Setting an underscore before capital and first letters
starts <- purrr::pmap(list(camelcases,
small_start,
capitals),
function(x,y,z)
if (length(z) == 0) {"_"} else {
c("_", paste0("_", z))
}
)
# split the strings by their capital letter sequences.
rests <- stringr::str_split(camelcases, "[A-Z]+")
# setting all peaces together:
# - pasting first and capital letters with the rest of the string
# - applying tolower, remove more than one "_" and starting "_"
corrected <- purrr::map2_chr(starts, rests, stringr::str_c, collapse = "") %>%
purrr::map_chr(stringr::str_to_lower) %>%
purrr::map_chr(~ stringr::str_replace_all(.x, "_+", "_")) %>%
purrr::map_chr(~ stringr::str_replace_all(.x, "^_+|_+$", ""))
corrected
}
camelCases <- c("smallCamelCase", "BigCamelCase", "mixed_Case", "snake_case",
"_camel_case__")
to_snake_case(camelCases)
#> [1] "small_camel_case" "big_camel_case" "mixed_case"
#> [4] "snake_case" "camel_case"
to_snake_case(names(iris))
#> [1] "sepal_length" "sepal_width" "petal_length" "petal_width"
#> [5] "species"
to_snake_case(c(".", "_"))
#> [1] "" ""
to_snake_case(NA)
#> [1] NA
to_snake_case(NULL)
#> character(0)
to_snake_case(character(0))
#> character(0)
to_snake_case(" N A")
#> [1] "n_a"
to_snake_case("LETTERS LETTERS")
#> [1] "letters_letters"
to_snake_case(LETTERS)
#> [1] "a" "b" "c" "d" "e" "f" "g" "h" "i" "j" "k" "l" "m" "n" "o" "p" "q"
#> [18] "r" "s" "t" "u" "v" "w" "x" "y" "z"Task:: Change column names from snake_case to camelCase.
Task: Mutate/Change/reorder columns depending on the row number (for example the last row).
iris %>% head(1) %>% dplyr::select(Species, dplyr::everything())
#> Species Sepal.Length Sepal.Width Petal.Length Petal.Width
#> 1 setosa 5.1 3.5 1.4 0.2
# ____________________________________________________________________________
setNames(iris[1, ], c("Species", setdiff(names(iris), "Species")))
#> Species Sepal.Length Sepal.Width Petal.Length Petal.Width
#> 1 5.1 3.5 1.4 0.2 setosaTask Work on more than two tables (Reduce())
Task Convert rownames of a data frame into first column.
4.5 Joins
Task: Join on two columns.