coalesce_one_hot function

Coalesce one-hot encoding back to a single variable

Helper function to reverse one-hot encoding post-imputation.

coalesce_one_hot(X, var_name, fast = TRUE)

Arguments

• X: A data.frame, data.table or matrix, for a single variable
• var_name: A character string, with the original variable label
• fast: Boolean, indicating whether to choose category with highest predicted probability (TRUE), or use predicted probabilities as weights in draw from random distribution

Returns

A vector of length equal to nrow(X), containing categorical labels corresponding to the columns of X