K-fold cross-validation over items

Splits items into k folds, fits on training items, evaluates on held-out items.

cv_items(transmatrix, k = 5L, priors = NULL, seed = NULL)

Arguments

transmatrix

numeric matrix from multi_transmat()

k

integer number of folds

priors

optional numeric vector of starting parameters

seed

optional integer random seed

Value

list with fold_results, mean_ll, total_ll, perplexity, se