rsample (version 1.1.0)

group_vfold_cv: Group V-Fold Cross-Validation

Description

Group V-fold cross-validation creates splits of the data based on some grouping variable (which may have more than a single row associated with it). The function can create as many splits as there are unique values of the grouping variable or it can create a smaller set of splits where more than one group is left out at a time. A common use of this kind of resampling is when you have repeated measures of the same subject.

Usage

group_vfold_cv(
  data,
  group = NULL,
  v = NULL,
  repeats = 1,
  balance = c("groups", "observations"),
  ...
)

Value

A tibble with classes group_vfold_cv, rset, tbl_df, tbl, and data.frame. The results include a column for the data split objects and an identification variable.

Arguments

data

A data frame.

group

A variable in data (single character or name) used for grouping observations with the same value to either the analysis or assessment set within a fold.

v

The number of partitions of the data set. If left as NULL (the default), v will be set to the number of unique values in the grouping variable, creating "leave-one-group-out" splits.

repeats

The number of times to repeat the V-fold partitioning.

balance

If v is less than the number of unique groups, how should groups be combined into folds? Should be one of "groups", which will assign roughly the same number of groups to each fold, or "observations", which will assign roughly the same number of observations to each fold.

...

Not currently used.

Examples

data(ames, package = "modeldata")

set.seed(123)
group_vfold_cv(ames, group = Neighborhood, v = 5)
group_vfold_cv(
  ames,
  group = Neighborhood,
  v = 5,
  balance = "observations"
)
group_vfold_cv(ames, group = Neighborhood, v = 5, repeats = 2)

# Leave-one-group-out CV
group_vfold_cv(ames, group = Neighborhood)