Title: Apply Functions to All Combinations of List Elements
Version: 0.4.2
Description: Provides an extension to the 'purrr' family of mapping functions to apply a function to each combination of elements in a list of inputs. Also includes functions for automatically detecting output type in mapping functions, finding every combination of elements of lists or rows of data frames, and applying multiple models to multiple subsets of a dataset.
License: MIT + file LICENSE
URL: https://crossmap.rossellhayes.com, https://github.com/rossellhayes/crossmap
BugReports: https://github.com/rossellhayes/crossmap/issues
Imports: backports, cli, dplyr (≥ 1.0.0), generics, lifecycle, purrr, rlang, stats, utils, vctrs
Suggests: broom, covr, crayon, estimatr, furrr, future, testthat, tibble, withr
Encoding: UTF-8
RoxygenNote: 7.3.2
NeedsCompilation: no
Packaged: 2025-04-24 15:14:51 UTC; alex
Author: Alexander Rossell Hayes ORCID iD [aut, cre, cph]
Maintainer: Alexander Rossell Hayes <alexander@rossellhayes.com>
Repository: CRAN
Date/Publication: 2025-04-25 08:50:06 UTC

crossmap: Apply Functions to All Combinations of List Elements

Description

logo

Provides an extension to the 'purrr' family of mapping functions to apply a function to each combination of elements in a list of inputs. Also includes functions for automatically detecting output type in mapping functions, finding every combination of elements of lists or rows of data frames, and applying multiple models to multiple subsets of a dataset.

Author(s)

Maintainer: Alexander Rossell Hayes alexander@rossellhayes.com (ORCID) [copyright holder]

See Also

Useful links:

Automatically generate names for vectors

Description

Automatically generate names for vectors

Usage

autonames(x, ..., trimws = TRUE)

Arguments

x

A vector

...

Additional arguments passed to format()

trimws

Whether to trim whitespace surrounding automatically formatted names. Defaults to TRUE.

Value

Returns the names of a named vector and the elements of an unnamed vector formatted as characters.

Examples

autonames(c(a = "apple", b = "banana", c = "cantaloupe"))
autonames(c("apple", "banana", "cantaloupe"))

autonames(10^(1:4))
autonames(10^(1:4), big.mark = ",")
autonames(10^(1:4), scientific = TRUE)

Cross map a model across multiple formulas, subsets, and weights

Description

Applies a modeling function to every combination of a set of formulas and a set of data subsets.

Usage

cross_fit(
  data,
  formulas,
  cols = NULL,
  weights = NULL,
  clusters = NULL,
  families = NULL,
  fn = lm,
  fn_args = list(),
  tidy = tidy_glance,
  tidy_args = list(),
  errors = c("stop", "warn")
)

Arguments

data

A data frame

formulas

A list of formulas to apply to each subset of the data. If named, these names will be used in the model column of the output. Otherwise, the formulas will be converted to strings in the model column.

cols

Columns to subset the data. Can be any expression supported by <tidy-select>. If NULL, the data is not subset into columns. Defaults to NULL.

weights

A list of columns passed to weights in fn. If one of the elements is NULL or NA, that model will not be weighted. Defaults to NULL.

clusters

A list of columns passed to clusters if supported by fn. If one of the elements is NULL or NA, that model will not be clustered. Defaults to NULL.

families

A list of glm model families passed to family if supported by fn. Defaults to gaussian("identity"), the equivalent of lm(). See family for examples.

fn

The modeling function. Either an unquoted function name or a purrr-style lambda function with two arguments. To use multiple modeling functions, see cross_fit_glm(). Defaults to lm.

fn_args

A list of additional arguments to fn.

tidy

A logical or function to use to tidy model output into data.frame columns. If TRUE, uses the default tidying function: tidy_glance(). If FALSE, NA, or NULL, the untidied model output will be returned in a list column named fit. An alternative function can be specified with an unquoted function name or a purrr-style lambda function with one argument (see usage with broom::tidy(conf.int = TRUE) in examples). Defaults to tidy_glance.

tidy_args

A list of additional arguments to the tidy function

errors

If "stop", the default, the function will stop and return an error if any subset produces an error. If "warn", the function will produce a warning for subsets that produce an error and return results for all subsets that do not.

Value

A tibble with a column for the model formula, columns for subsets, columns for the model family and type (if applicable), columns for the weights and clusters (if applicable), and columns of tidy model output or a list column of models (if tidy = FALSE)

See Also

cross_fit_glm() to map a model across multiple model types.

cross_fit_robust() to map robust linear models.

xmap() to apply any function to combinations of inputs.

Examples

cross_fit(mtcars, mpg ~ wt, cyl)
cross_fit(mtcars, list(mpg ~ wt, mpg ~ hp), cyl)
cross_fit(mtcars, list(wt = mpg ~ wt, hp = mpg ~ hp), cyl)

cross_fit(mtcars, list(mpg ~ wt, mpg ~ hp), c(cyl, vs))
cross_fit(mtcars, list(mpg ~ wt, mpg ~ hp), dplyr::starts_with("c"))

cross_fit(mtcars, list(hp = mpg ~ hp), cyl, weights = wt)
cross_fit(mtcars, list(hp = mpg ~ hp), cyl, weights = c(wt, NA))

cross_fit(
  mtcars, list(vs ~ cyl, vs ~ hp), am,
  fn = glm, fn_args = list(family = binomial(link = logit))
)
cross_fit(
  mtcars, list(vs ~ cyl, vs ~ hp), am,
  fn = ~ glm(.x, .y, family = binomial(link = logit))
)

cross_fit(mtcars, list(mpg ~ wt, mpg ~ hp), cyl, tidy = FALSE)
cross_fit(mtcars, list(mpg ~ wt, mpg ~ hp), cyl, tidy_args = c(conf.int = TRUE))

cross_fit(mtcars, list(mpg ~ wt, mpg ~ hp), cyl, tidy = broom::tidy)
cross_fit(
  mtcars, list(mpg ~ wt, mpg ~ hp), cyl,
  tidy = ~ broom::tidy(., conf.int = TRUE)
)

Cross fit generalized linear models

Description

Cross fit generalized linear models

Usage

cross_fit_glm(
  data,
  formulas,
  cols = NULL,
  weights = NULL,
  families = gaussian(link = identity),
  fn_args = list(),
  tidy = tidy_glance,
  tidy_args = list(),
  errors = c("stop", "warn")
)

Arguments

data

A data frame

formulas

A list of formulas to apply to each subset of the data. If named, these names will be used in the model column of the output. Otherwise, the formulas will be converted to strings in the model column.

cols

Columns to subset the data. Can be any expression supported by <tidy-select>. If NULL, the data is not subset into columns. Defaults to NULL.

weights

A list of columns passed to weights in fn. If one of the elements is NULL or NA, that model will not be weighted. Defaults to NULL.

families

A list of glm model families. Defaults to gaussian("identity"), the equivalent of lm(). See family for examples.

fn_args

A list of additional arguments to glm().

tidy

A logical or function to use to tidy model output into data.frame columns. If TRUE, uses the default tidying function: tidy_glance(). If FALSE, NA, or NULL, the untidied model output will be returned in a list column named fit. An alternative function can be specified with an unquoted function name or a purrr-style lambda function with one argument (see usage with broom::tidy(conf.int = TRUE) in examples). Defaults to tidy_glance.

tidy_args

A list of additional arguments to the tidy function

errors

If "stop", the default, the function will stop and return an error if any subset produces an error. If "warn", the function will produce a warning for subsets that produce an error and return results for all subsets that do not.

Value

A tibble with a column for the model formula, columns for subsets, columns for the model family and type, columns for the weights (if applicable), and columns of tidy model output or a list column of models (if tidy = FALSE)

See Also

cross_fit() to use any modeling function.

Examples

cross_fit_glm(
  data     = mtcars,
  formulas = list(am ~ gear, am ~ cyl),
  cols     = vs,
  families = list(gaussian("identity"), binomial("logit"))
)

Cross fit robust linear models

Description

Cross fit robust linear models

Usage

cross_fit_robust(
  data,
  formulas,
  cols = NULL,
  weights = NULL,
  clusters = NULL,
  fn_args = list(),
  tidy = tidy_glance,
  tidy_args = list(),
  errors = c("stop", "warn")
)

Arguments

data

A data frame

formulas

A list of formulas to apply to each subset of the data. If named, these names will be used in the model column of the output. Otherwise, the formulas will be converted to strings in the model column.

cols

Columns to subset the data. Can be any expression supported by <tidy-select>. If NULL, the data is not subset into columns. Defaults to NULL.

weights

A list of columns passed to weights in fn. If one of the elements is NULL or NA, that model will not be weighted. Defaults to NULL.

clusters

A list of columns passed to clusters. If one of the elements is NULL or NA, that model will not be clustered. Defaults to NULL.

fn_args

A list of additional arguments to estimatr::lm_robust().

tidy

A logical or function to use to tidy model output into data.frame columns. If TRUE, uses the default tidying function: tidy_glance(). If FALSE, NA, or NULL, the untidied model output will be returned in a list column named fit. An alternative function can be specified with an unquoted function name or a purrr-style lambda function with one argument (see usage with broom::tidy(conf.int = TRUE) in examples). Defaults to tidy_glance.

tidy_args

A list of additional arguments to the tidy function

errors

If "stop", the default, the function will stop and return an error if any subset produces an error. If "warn", the function will produce a warning for subsets that produce an error and return results for all subsets that do not.

Value

A tibble with a column for the model formula, columns for subsets, columns for the weights and clusters (if applicable), and columns of tidy model output or a list column of models (if tidy = FALSE)

See Also

cross_fit() to use any modeling function.

Examples


cross_fit_robust(mtcars, mpg ~ wt, cyl, clusters = list(NULL, am))

Crossing join

Description

Adds columns from a set of data frames, creating all combinations of their rows

Usage

cross_join(..., copy = FALSE)

Arguments

...

Data frames or a list of data frames – including data frame extensions (e.g. tibbles) and lazy data frames (e.g. from dbplyr or dtplyr). NULL inputs are silently ignored.

copy

If inputs are not from the same data source, and copy is TRUE, then they will be copied into the same src as the first input. This allows you to join tables across srcs, but it is a potentially expensive operation so you must opt into it.

Value

An object of the same type as the first input. The order of the rows and columns of the first input is preserved as much as possible. The output has the following properties:

See Also

cross_list() to find combinations of elements of vectors and lists.

Examples

fruits <- dplyr::tibble(
  fruit = c("apple", "banana", "cantaloupe"),
  color = c("red", "yellow", "orange")
)

desserts <- dplyr::tibble(
  dessert = c("cupcake", "muffin", "streudel"),
  makes   = c(8, 6, 1)
)

cross_join(fruits, desserts)
cross_join(list(fruits, desserts))
cross_join(rep(list(fruits), 3))

List all combinations of values

Description

List all combinations of values

Usage

cross_list(...)

cross_tbl(...)

Arguments

...

Inputs or a list of inputs. NULL inputs are silently ignored.

Value

A list for cross_list() or tibble for cross_tbl(). Names will match the names of the inputs. Unnamed inputs will be left unnamed for cross_list() and automatically named for cross_tbl().

See Also

cross_join() to find combinations of data frame rows.

purrr::cross() for an implementation that results in a differently formatted list.

expand.grid() for an implementation that results in a data.frame.

Examples

fruits   <- c("apple", "banana", "cantaloupe")
desserts <- c("cupcake", "muffin", "streudel")

cross_list(list(fruits, desserts))
cross_list(fruits, desserts)
cross_tbl(fruits, desserts)

cross_list(list(fruit = fruits, dessert = desserts))
cross_list(fruit = fruits, dessert = desserts)
cross_tbl(fruit = fruits, dessert = desserts)

Parallelized mapping functions that automatically determine type

Description

These functions work exactly the same as map_vec(), map2_vec(), pmap_vec(), imap_vec() and xmap_vec(), but allow you to map in parallel.

Usage

future_map_vec(
  .x,
  .f,
  ...,
  .class = NULL,
  .progress = FALSE,
  .options = furrr::furrr_options()
)

future_map2_vec(
  .x,
  .y,
  .f,
  ...,
  .class = NULL,
  .progress = FALSE,
  .options = furrr::furrr_options()
)

future_pmap_vec(
  .l,
  .f,
  ...,
  .class = NULL,
  .progress = FALSE,
  .options = furrr::furrr_options()
)

future_imap_vec(
  .x,
  .f,
  ...,
  .class = NULL,
  .progress = FALSE,
  .options = furrr::furrr_options()
)

future_xmap_vec(
  .l,
  .f,
  ...,
  .class = NULL,
  .progress = FALSE,
  .options = furrr::furrr_options()
)

Arguments

.x

A list or atomic vector.

.f

A function, formula, or vector (not necessarily atomic).

If a function, it is used as is.

If a formula, e.g. ~ .x + 2, it is converted to a function. There are three ways to refer to the arguments:

  • For a single argument function, use .

  • For a two argument function, use .x and .y

  • For more arguments, use ..1, ..2, ..3 etc

This syntax allows you to create very compact anonymous functions.

If character vector, numeric vector, or list, it is converted to an extractor function. Character vectors index by name and numeric vectors index by position; use a list to index by position and name at different levels. If a component is not present, the value of .default will be returned.

...

Additional arguments passed on to .f

.class

If .class is specified, all

.progress

A single logical. Should a progress bar be displayed? Only works with multisession, multicore, and multiprocess futures. Note that if a multicore/multisession future falls back to sequential, then a progress bar will not be displayed.

Warning: The .progress argument will be deprecated and removed in a future version of furrr in favor of using the more robust progressr package.

.options

The future specific options to use with the workers. This must be the result from a call to furrr_options().

.y

A vector the same length as .x. Vectors of length 1 will be recycled.

.l

A list of vectors, such as a data frame. The length of .l determines the number of arguments that .f will be called with. List names will be used if present.

Value

Equivalent to map_vec(), map2_vec(), pmap_vec(), imap_vec() and xmap_vec()

See Also

The original functions: furrr::future_map(), furrr::future_map2(), furrr::future_pmap(), furrr::future_imap() and future_xmap()

Non-parallelized equivalents: map_vec(), map2_vec(), pmap_vec(), imap_vec() and xmap_vec()

Examples

fruits   <- c("apple", "banana", "carrot", "durian", "eggplant")
desserts <- c("bread", "cake", "cupcake", "streudel", "muffin")
x        <- sample(5)
y        <- sample(5)
z        <- sample(5)
names(z) <- fruits

future_map_vec(x, ~ . ^ 2)
future_map_vec(fruits, paste0, "s")

future_map2_vec(x, y, ~ .x + .y)
future_map2_vec(fruits, desserts, paste)

future_pmap_vec(list(x, y, z), sum)
future_pmap_vec(list(x, fruits, desserts), paste)

future_imap_vec(x, ~ .x + .y)
future_imap_vec(x, ~ paste0(.y, ": ", .x))
future_imap_vec(z, paste)

future_xmap_vec(list(x, y), ~ .x * .y)
future_xmap_vec(list(fruits, desserts), paste)

Map over each combination of list elements simultaneously via futures

Description

These functions work exactly the same as xmap() functions, but allow you to run the map in parallel using future::future()

Usage

future_xmap(.l, .f, ..., .progress = FALSE, .options = furrr::furrr_options())

future_xmap_chr(
  .l,
  .f,
  ...,
  .progress = FALSE,
  .options = furrr::furrr_options()
)

future_xmap_dbl(
  .l,
  .f,
  ...,
  .progress = FALSE,
  .options = furrr::furrr_options()
)

future_xmap_dfc(
  .l,
  .f,
  ...,
  .progress = FALSE,
  .options = furrr::furrr_options()
)

future_xmap_dfr(
  .l,
  .f,
  ...,
  .id = NULL,
  .progress = FALSE,
  .options = furrr::furrr_options()
)

future_xmap_int(
  .l,
  .f,
  ...,
  .progress = FALSE,
  .options = furrr::furrr_options()
)

future_xmap_lgl(
  .l,
  .f,
  ...,
  .progress = FALSE,
  .options = furrr::furrr_options()
)

future_xmap_raw(
  .l,
  .f,
  ...,
  .progress = FALSE,
  .options = furrr::furrr_options()
)

future_xwalk(.l, .f, ..., .progress = FALSE, .options = furrr::furrr_options())

Arguments

.l

A list of vectors, such as a data frame. The length of .l determines the number of arguments that .f will be called with. List names will be used if present.

.f

A function, formula, or vector (not necessarily atomic).

If a function, it is used as is.

If a formula, e.g. ~ .x + 2, it is converted to a function. There are three ways to refer to the arguments:

  • For a single argument function, use .

  • For a two argument function, use .x and .y

  • For more arguments, use ..1, ..2, ..3 etc

This syntax allows you to create very compact anonymous functions.

If character vector, numeric vector, or list, it is converted to an extractor function. Character vectors index by name and numeric vectors index by position; use a list to index by position and name at different levels. If a component is not present, the value of .default will be returned.

...

Additional arguments passed on to .f

.progress

A single logical. Should a progress bar be displayed? Only works with multisession, multicore, and multiprocess futures. Note that if a multicore/multisession future falls back to sequential, then a progress bar will not be displayed.

Warning: The .progress argument will be deprecated and removed in a future version of furrr in favor of using the more robust progressr package.

.options

The future specific options to use with the workers. This must be the result from a call to furrr_options().

.id

Either a string or NULL. If a string, the output will contain a variable with that name, storing either the name (if .x is named) or the index (if .x is unnamed) of the input. If NULL, the default, no variable will be created.

Only applies to ⁠_dfr⁠ variant.

Value

An atomic vector, list, or data frame, depending on the suffix. Atomic vectors and lists will be named if the first element of .l is named.

If all input is length 0, the output will be length 0. If any input is length 1, it will be recycled to the length of the longest.

See Also

xmap() to run functions without parallel processing.

future_xmap_vec() to automatically determine output type.

future_xmap_mat() and future_xmap_arr() to return results in a matrix or array.

furrr::future_map(), furrr::future_map2(), and furrr::future_pmap() for other parallelized mapping functions.

Examples

future_xmap(list(1:5, 1:5), ~ .y * .x)
future_xmap_dbl(list(1:5, 1:5), ~ .y * .x)
future_xmap_chr(list(1:5, 1:5), ~ paste(.y, "*", .x, "=", .y * .x))

apples_and_bananas <- list(
  x = c("apples", "bananas"),
  pattern = "a",
  replacement = c("oo", "ee")
)

future_xmap_chr(apples_and_bananas, gsub)

formulas <- list(mpg ~ wt, mpg ~ hp)
subsets  <- split(mtcars, mtcars$cyl)

future_xmap(list(subsets, formulas), ~ lm(.y, data = .x))

Parallelized cross map returning a matrix or array

Description

Parallelized cross map returning a matrix or array

Usage

future_xmap_mat(
  .l,
  .f,
  ...,
  .names = TRUE,
  .progress = FALSE,
  .options = furrr::furrr_options()
)

future_xmap_arr(
  .l,
  .f,
  ...,
  .names = TRUE,
  .progress = FALSE,
  .options = furrr::furrr_options()
)

Arguments

.l

A list of vectors, such as a data frame. The length of .l determines the number of arguments that .f will be called with. List names will be used if present.

.f

A function, formula, or vector (not necessarily atomic).

If a function, it is used as is.

If a formula, e.g. ~ .x + 2, it is converted to a function. There are three ways to refer to the arguments:

  • For a single argument function, use .

  • For a two argument function, use .x and .y

  • For more arguments, use ..1, ..2, ..3 etc

This syntax allows you to create very compact anonymous functions.

If character vector, numeric vector, or list, it is converted to an extractor function. Character vectors index by name and numeric vectors index by position; use a list to index by position and name at different levels. If a component is not present, the value of .default will be returned.

...

Additional arguments passed on to .f

.names

A logical indicating whether to give names to the dimensions of the matrix or array. If inputs are named, the names are used. If inputs are unnamed, the elements of the input are used as names. Defaults to TRUE.

.progress

A single logical. Should a progress bar be displayed? Only works with multisession, multicore, and multiprocess futures. Note that if a multicore/multisession future falls back to sequential, then a progress bar will not be displayed.

Warning: The .progress argument will be deprecated and removed in a future version of furrr in favor of using the more robust progressr package.

.options

The future specific options to use with the workers. This must be the result from a call to furrr_options().

Value

A matrix (for future_xmap_mat()) or array (for future_xmap_arr()) with dimensions matching the lengths of each input in .l.

See Also

Unparallelized versions: xmap_mat() and xmap_arr()

future_xmap_vec() to return a vector.

future_xmap() for the underlying functions.

Examples

future_xmap_mat(list(1:3, 1:3),  ~ ..1 * ..2)

fruits <- c(a = "apple", b = "banana", c = "cantaloupe")
future_xmap_mat(list(1:3, fruits), paste)
future_xmap_mat(list(1:3, fruits), paste, .names = FALSE)

future_xmap_arr(list(1:3, 1:3, 1:3),  ~ ..1 * ..2 * ..3)

Mapping functions that automatically determine type

Description

These functions work exactly the same as typed variants of purrr::map(), purrr::map2(), purrr::pmap(), purrr::imap() and xmap() (e.g. purrr::map_chr()), but automatically determine the type.

Usage

map_vec(.x, .f, ..., .class = NULL)

map2_vec(.x, .y, .f, ..., .class = NULL)

pmap_vec(.l, .f, ..., .class = NULL)

imap_vec(.x, .f, ..., .class = NULL)

xmap_vec(.l, .f, ..., .class = NULL)

Arguments

.x

A list or atomic vector.

.f

A function, formula, or vector (not necessarily atomic).

If a function, it is used as is.

If a formula, e.g. ~ .x + 2, it is converted to a function. There are three ways to refer to the arguments:

  • For a single argument function, use .

  • For a two argument function, use .x and .y

  • For more arguments, use ..1, ..2, ..3 etc

This syntax allows you to create very compact anonymous functions.

If character vector, numeric vector, or list, it is converted to an extractor function. Character vectors index by name and numeric vectors index by position; use a list to index by position and name at different levels. If a component is not present, the value of .default will be returned.

...

Additional arguments passed on to .f

.class

If .class is specified, all

.y

A vector the same length as .x. Vectors of length 1 will be recycled.

.l

A list of vectors, such as a data frame. The length of .l determines the number of arguments that .f will be called with. List names will be used if present.

Value

Equivalent to the typed variants of purrr::map(), purrr::map2(), purrr::pmap(), purrr::imap() and xmap() with the type automatically determined.

If the output contains multiple types, the type is determined from the highest type of the components in the hierarchy raw < logical < integer < double < complex < character < list (as in c()).

If the output contains elements that cannot be coerced to vectors (e.g. lists), the output will be a list.

See Also

The original functions: purrr::map(), purrr::map2(), purrr::pmap(), purrr::imap() and xmap()

Parallelized equivalents: future_map_vec(), future_map2_vec(), future_pmap_vec(), future_imap_vec() and future_xmap_vec()

Examples

fruits   <- c("apple", "banana", "cantaloupe", "durian", "eggplant")
desserts <- c("bread", "cake", "cupcake", "muffin", "streudel")
x        <- sample(5)
y        <- sample(5)
z        <- sample(5)
names(z) <- fruits

map_vec(x, ~ . ^ 2)
map_vec(fruits, paste0, "s")

map2_vec(x, y, ~ .x + .y)
map2_vec(fruits, desserts, paste)

pmap_vec(list(x, y, z), sum)
pmap_vec(list(x, fruits, desserts), paste)

imap_vec(x, ~ .x + .y)
imap_vec(x, ~ paste0(.y, ": ", .x))
imap_vec(z, paste)

xmap_vec(list(x, y), ~ .x * .y)
xmap_vec(list(fruits, desserts), paste)

Turn an object into a tidy tibble with glance information

Description

Apply both generics::tidy() and generics::glance() to an object and return a single tibble with both sets of information.

Usage

tidy_glance(x, ..., tidy_args = list(), glance_args = list())

Arguments

x

An object to be converted into a tidy tibble.

...

Additional arguments passed to generics::tidy() and generics::glance().

Arguments are passed to both methods, but should be ignored by the inapplicable method. For example, if called on an lm object, conf.int will affect generics::tidy() but not generics::glance().

tidy_args

A list of additional arguments passed only to generics::tidy().

glance_args

A list of additional arguments passed only to generics::glance().

Value

A tibble with columns and rows from generics::tidy() and columns of repeated rows from generics::glance().

Column names that appear in both the tidy data and glance data will be disambiguated by appending "model." to the glance column names.

Examples

mod <- lm(mpg ~ wt + qsec, data = mtcars)
tidy_glance(mod)
tidy_glance(mod, conf.int = TRUE)
tidy_glance(mod, tidy_args = list(conf.int = TRUE))

Map over each combination of list elements

Description

These functions are variants of purrr::pmap() that iterate over each combination of elements in a list.

Usage

xmap(.l, .f, ...)

xmap_chr(.l, .f, ...)

xmap_dbl(.l, .f, ...)

xmap_dfc(.l, .f, ...)

xmap_dfr(.l, .f, ..., .id = NULL)

xmap_int(.l, .f, ...)

xmap_lgl(.l, .f, ...)

xmap_raw(.l, .f, ...)

xwalk(.l, .f, ...)

Arguments

.l

A list of vectors, such as a data frame. The length of .l determines the number of arguments that .f will be called with. List names will be used if present.

.f

A function, formula, or vector (not necessarily atomic).

If a function, it is used as is.

If a formula, e.g. ~ .x + 2, it is converted to a function. There are three ways to refer to the arguments:

  • For a single argument function, use .

  • For a two argument function, use .x and .y

  • For more arguments, use ..1, ..2, ..3 etc

This syntax allows you to create very compact anonymous functions.

If character vector, numeric vector, or list, it is converted to an extractor function. Character vectors index by name and numeric vectors index by position; use a list to index by position and name at different levels. If a component is not present, the value of .default will be returned.

...

Additional arguments passed on to .f

.id

Either a string or NULL. If a string, the output will contain a variable with that name, storing either the name (if .x is named) or the index (if .x is unnamed) of the input. If NULL, the default, no variable will be created.

Only applies to ⁠_dfr⁠ variant.

Details

Typed variants return a vector of the specified type. To automatically determine type, try xmap_vec().

To return results as a matrix or array, try xmap_mat() and xmap_arr().

Note that a data frame is a very important special case, in which case xmap() and xwalk() apply the function .f to each row. xmap_dfr() and xmap_dfc() return data frames created by row-binding and column-binding respectively.

Value

An atomic vector, list, or data frame, depending on the suffix. Atomic vectors and lists will be named if the first element of .l is named.

If all input is length 0, the output will be length 0. If any input is length 1, it will be recycled to the length of the longest.

See Also

xmap_vec() to automatically determine output type.

xmap_mat() and xmap_arr() to return results in a matrix or array.

future_xmap() to run xmap functions with parallel processing.

cross_fit() to apply multiple models to multiple subsets of data.

cross_list() to find combinations of list elements.

purrr::map(), purrr::map2(), and purrr::pmap() for other mapping functions.

Examples

xmap(list(1:5, 1:5), ~ .y * .x)
xmap_dbl(list(1:5, 1:5), ~ .y * .x)
xmap_chr(list(1:5, 1:5), ~ paste(.y, "*", .x, "=", .y * .x))

apples_and_bananas <- list(
  x = c("apples", "bananas"),
  pattern = "a",
  replacement = c("oo", "ee")
)

xmap_chr(apples_and_bananas, gsub)

formulas <- list(mpg ~ wt, mpg ~ hp)
subsets  <- split(mtcars, mtcars$cyl)

xmap(list(subsets, formulas), ~ lm(.y, data = .x))
xmap(list(data = subsets, formula = formulas), lm)

Return a table applying a function to all combinations of list elements

Description

Return a table applying a function to all combinations of list elements

Usage

xmap_mat(.l, .f, ..., .names = TRUE)

xmap_arr(.l, .f, ..., .names = TRUE)

Arguments

.l

A list of vectors, such as a data frame. The length of .l determines the number of arguments that .f will be called with. List names will be used if present.

.f

A function, formula, or vector (not necessarily atomic).

If a function, it is used as is.

If a formula, e.g. ~ .x + 2, it is converted to a function. There are three ways to refer to the arguments:

  • For a single argument function, use .

  • For a two argument function, use .x and .y

  • For more arguments, use ..1, ..2, ..3 etc

This syntax allows you to create very compact anonymous functions.

If character vector, numeric vector, or list, it is converted to an extractor function. Character vectors index by name and numeric vectors index by position; use a list to index by position and name at different levels. If a component is not present, the value of .default will be returned.

...

Additional arguments passed on to .f

.names

A logical indicating whether to give names to the dimensions of the matrix or array. If inputs are named, the names are used. If inputs are unnamed, the elements of the input are used as names. Defaults to TRUE.

Value

A matrix (for xmap_mat()) or array (for xmap_arr()) with dimensions equal to the lengths of each input in .l.

See Also

future_xmap_mat() and future_xmap_arr() to run functions in parallel.

xmap_vec() to return a vector.

xmap() for the underlying functions.

Examples

xmap_mat(list(1:3, 1:3),  ~ ..1 * ..2)

fruits <- c(a = "apple", b = "banana", c = "cantaloupe")
xmap_mat(list(1:3, fruits), paste)
xmap_mat(list(1:3, fruits), paste, .names = FALSE)

xmap_arr(list(1:3, 1:3, 1:3),  ~ ..1 * ..2 * ..3)

Get one or more elements deep within a nested data structure

Description

xpluck() provides an alternative to purrr::pluck(). Unlike purrr::pluck(), xpluck() allows you to extract multiple indices at each nesting level.

Usage

xpluck(.x, ..., .default = NULL)

Arguments

.x

A list or vector

...

A list of accessors for indexing into the object. Can be positive integers, negative integers (to index from the right), strings (to index into names) or missing (to keep all elements at a given level).

Unlike purrr::pluck(), each accessor may be a vector to extract multiple elements.

If an accessor has length 0 (e.g. NULL, character(0) or numeric(0)), xpluck() will return NULL.

.default

Value to use if target is NULL or absent.

Value

A list or vector.

Examples

obj1 <- list("a", list(1, elt = "foo"))
obj2 <- list("b", list(2, elt = "bar"))
x <- list(obj1, obj2)

xpluck(x, 1:2, 2)
xpluck(x, , 2)

xpluck(x, , 2, 1)
xpluck(x, , 2, 2)
xpluck(x, , 2, 1:2)