Construct design-aware resamples

Usage

som_resamples(
  data,
  method = c("full", "subsample", "group_subsample", "block_subsample",
    "leave_domain_out", "custom"),
  repeats = 50L,
  prop = 0.8,
  seed = 1L,
  unit = "group",
  domain = "domain",
  splits = NULL
)

Arguments

data: A som_data object.
method: One of "full", "subsample", "group_subsample", "block_subsample", "leave_domain_out" or "custom".
repeats: Number of repeated subsamples.
prop: Proportion of rows or sampling units retained for analysis.
seed: Random seed used to generate all splits.
unit: Metadata column name or vector defining groups or predefined blocks. block_subsample does not infer blocks from row order.
domain: Metadata column name or vector defining transfer domains.
splits: For method = "custom", a list containing analysis and optional assessment row indices.

Value

A som_resamples object. The object records the number of distinct analysis sets in n_unique_analysis_splits and identifies repeated sets in duplicate_analysis_splits.

Examples

x <- matrix(seq_len(120), nrow = 30)
data <- som_data(x, group = rep(paste0("site_", 1:10), each = 3))
som_resamples(data, method = "group_subsample", repeats = 2, seed = 1)
#> <som_resamples>
#>   method    : group_subsample 
#>   splits    : 2 
#>   unique analysis sets: 2 
#>   analysis  : 24-24 rows
#>   assessment: 6-6 rows