One nesting level has a single observation and throws an error

[ericpgreen]

Thanks for creating this package. I think it will be a nice shortcut for an analysis I'm working on.

I'm getting an error because one nesting level in the dataset (of 800+) only has one observation. I take it that nested_resamples() needs at least two observations. Is this right? And if so, is there a work around?

>   split <- nested_resamples(nested_df, rsample::initial_split())
Error in `purrr::map()`:
ℹ In index: 1334.
Caused by error in `purrr::map()`:
ℹ In index: 1.
Caused by error in `rsplit()`:
! At least one row should be selected for the analysis set.
Backtrace:
  1. nestedmodels::nested_resamples(nested_df, rsample::initial_split())
 11. rsample::initial_split(data = `<tibble[,20]>`)
 12. rsample::mc_cv(...)
 13. rsample:::mc_splits(...)
 14. purrr::map(indices, make_splits, data = data, class = "mc_split")
 15. purrr:::map_("list", .x, .f, ..., .progress = .progress)
 20. rsample:::make_splits.list(.x[[i]], ...)
 21. rsample:::rsplit(data, x$analysis, x$assessment)