Add group_bootstraps()

[mikemahoney218]

This PR is based off of #315 .

This PR fixes #207 (and we're going to open a new issue for stratification with grouping, to get input from a wider audience).

This PR adds group_bootstraps(), patterned off group_mc_cv(), in order to provide a way to bootstrap with defined groups. In order to do that, it adds an argument replace to balance_prop(), which is TRUE for group_bootstraps() and FALSE for group_mc_cv().

library(rsample)

set.seed(2)
dat1 <- tibble::tibble(a = 1:20, b = letters[1:20], c = rep(1:4, 5))

group_bootstraps(dat1, c, times = 2)$splits[[1]] |> 
  analysis()
#> # A tibble: 20 × 3
#>        a b         c
#>    <int> <chr> <int>
#>  1     1 a         1
#>  2     1 a         1
#>  3     2 b         2
#>  4     4 d         4
#>  5     5 e         1
#>  6     5 e         1
#>  7     6 f         2
#>  8     8 h         4
#>  9     9 i         1
#> 10     9 i         1
#> 11    10 j         2
#> 12    12 l         4
#> 13    13 m         1
#> 14    13 m         1
#> 15    14 n         2
#> 16    16 p         4
#> 17    17 q         1
#> 18    17 q         1
#> 19    18 r         2
#> 20    20 t         4

group_bootstraps(dat1, c, times = 2)$splits[[1]] |> 
  analysis()
#> # A tibble: 20 × 3
#>        a b         c
#>    <int> <chr> <int>
#>  1     1 a         1
#>  2     2 b         2
#>  3     2 b         2
#>  4     3 c         3
#>  5     5 e         1
#>  6     6 f         2
#>  7     6 f         2
#>  8     7 g         3
#>  9     9 i         1
#> 10    10 j         2
#> 11    10 j         2
#> 12    11 k         3
#> 13    13 m         1
#> 14    14 n         2
#> 15    14 n         2
#> 16    15 o         3
#> 17    17 q         1
#> 18    18 r         2
#> 19    18 r         2
#> 20    19 s         3

group_bootstraps(dat1, c, times = 2)$splits[[1]] |> 
  assessment()
#> # A tibble: 5 × 3
#>       a b         c
#>   <int> <chr> <int>
#> 1     2 b         2
#> 2     6 f         2
#> 3    10 j         2
#> 4    14 n         2
#> 5    18 r         2

group_bootstraps(dat1, c, times = 2)$splits[[1]] |> 
  assessment()
#> # A tibble: 5 × 3
#>       a b         c
#>   <int> <chr> <int>
#> 1     1 a         1
#> 2     5 e         1
#> 3     9 i         1
#> 4    13 m         1
#> 5    17 q         1

^{Created on 2022-06-28 by the reprex package (v2.0.1)}

[mikemahoney218]

E> Cannot initiate the connection to ppa.launchpad.net:80

Looks like ubuntu checks are failing for network reasons

[juliasilge]

This is really nice! ⭐

[github-actions]

This pull request has been automatically locked. If you believe you have found a related problem, please file a new issue (with a reprex: https://reprex.tidyverse.org) and link to this issue.