cv_sparse_log_contrast: Perform cross validation for tuning parameter selection for...
In jacobbien/trac: Tree-based Aggregation of Compositional Data

This function is to be called after calling sparse_log_contrast. It performs nfold-fold cross validation.

cv_sparse_log_contrast(
  fit,
  Z,
  y,
  folds = NULL,
  nfolds = 5,
  summary_function = stats::median
)

`fit`	output of `sparse_log_contrast` function.
`Z, y`	same arguments as passed to `sparse_log_contrast`. C that is used will be taken from fit object.
`folds`	a partition of `1:nrow(Z)`.
`nfolds`	number of folds for cross-validation
`summary_function`	how to combine the errors calculated on each observation within a fold (e.g. mean or median)