Tidy, glance and augment methods for emeans fits

Methods for the generics::tidy(), generics::glance() and generics::augment() generics, which return plain data frames and need only the generics package to be installed.

Usage

# S3 method for class 'emeans'
tidy(x, ...)

# S3 method for class 'emeans'
glance(x, ...)

# S3 method for class 'emeans'
augment(x, data = NULL, newdata = NULL, ...)

Arguments

x: An "emeans" object.
...: Unused.
data: The data that was clustered, as a data frame or matrix. Used when newdata is NULL; needs a fit made with keep_cluster = TRUE.
newdata: Optional new observations, labelled with predict().

Value

tidy() returns one row per cluster with the center columns, size, withinss and cluster (a factor). glance() returns one row with totss, tot.withinss, betweenss, obj (the last pass's objective), iter, epochs and n_splits; the sums of squares are NA for a fit made with keep_cluster = FALSE, which also has no withinss column in tidy(). augment() returns the data with a .cluster factor column.

Examples

if (requireNamespace("generics", quietly = TRUE)) {
  set.seed(1)
  x <- data.frame(a = c(rnorm(100), rnorm(100, mean = 4)),
                  b = c(rnorm(100), rnorm(100, mean = 4)))
  fit <- emeans(x, 2, threads = 1)
  generics::tidy(fit)
  generics::glance(fit)
  head(generics::augment(fit, data = x))
}
#>            a          b .cluster
#> 1 -0.6264538  0.4094018        1
#> 2  0.1836433  1.6888733        1
#> 3 -0.8356286  1.5865884        1
#> 4  1.5952808 -0.3309078        1
#> 5  0.3295078 -2.2852355        1
#> 6 -0.8204684  2.4976616        1