rset_reconstruct: Extending rsample with new rset subclasses

In tidymodels/rsample: General Resampling Infrastructure

Extending rsample with new rset subclasses

Description

rset_reconstruct() encapsulates the logic for allowing new rset subclasses to work properly with vctrs (through vctrs::vec_restore()) and dplyr (through dplyr::dplyr_reconstruct()). It is intended to be a developer tool, and is not required for normal usage of rsample.

Usage

rset_reconstruct(x, to)

Arguments

x	A data frame to restore to an rset subclass.
to	An rset subclass to restore to.

Details

rset objects are considered "reconstructable" after a vctrs/dplyr operation if:

• x and to both have an identical column named "splits" (column and row order do not matter).

• x and to both have identical columns prefixed with "id" (column and row order do not matter).

Value

x restored to the rset subclass of to.

Examples

to <- bootstraps(mtcars, times = 25)

# Imitate a vctrs/dplyr operation,
# where the class might be lost along the way
x <- tibble::as_tibble(to)

# Say we added a new column to `x`. Here we mock a `mutate()`.
x$foo <- "bar"

# This is still reconstructable to `to`
rset_reconstruct(x, to)

# Say we lose the first row
x <- x[-1, ]

# This is no longer reconstructable to `to`, as `x` is no longer an rset
# bootstraps object with 25 bootstraps if one is lost!
rset_reconstruct(x, to)

tidymodels/rsample documentation built on Sept. 29, 2024, 10:48 p.m.