impute: Regular imputation

In emil: Evaluation of Modeling without Information Leakage

Description

If you want to impute, build model and predict you should use pre_impute_median or pre_impute_knn. This function imputes using all observations without caring about cross-validation folds.

Usage

impute_knn(x, k = 0.05, distance_matrix = "auto")

impute_median(x)

Arguments

x	Dataset.
k	Number of nearest neighbors to use.
distance_matrix	Distance matrix.

Details

For additional information on the parameters see pre_impute_knn and pre_impute.

Value

An imputed matrix.

Author(s)

Christofer Bäcklin

See Also

emil, pre_process, pre_impute_knn, pre_impute_median

Examples

x <- matrix(rnorm(36), 6, 6)
x[sample(length(x), 5)] <- NA
impute_knn(x)
impute_median(x)

emil documentation built on Aug. 1, 2018, 1:03 a.m.