Feature imputation using random forests neighborhoods

Source: R/neighborhood_imputation.R

This function uses the neighborhoods implied by a random forest to impute missing features. The neighbors of a data point are all the training points assigned to the same leaf in at least one tree in the forest. The weight of each neighbor is the fraction of trees in the forest for which it was assigned to the same leaf. We impute a missing feature for a point by computing the weighted average feature value, using neighborhood weights, using all of the point's neighbors.

impute_features(
  object,
  newdata,
  seed = round(runif(1) * 10000),
  use_mean_imputation_fallback = FALSE
)

Arguments

object

an object of class `forestry`

newdata

the feature data.frame we will impute missing features for.

seed

a random seed passed to the predict method of forestry

use_mean_imputation_fallback

if TRUE, mean imputation (for numeric variables) and mode imputation (for factor variables) is used for missing features for which all neighbors also had the corresponding feature missing; if FALSE these missing features remain NAs in the data frame returned by `impute_features`.

Value

A data.frame that is newdata with imputed missing values.

Examples

iris_with_missing <- iris
idx_miss_factor <- sample(nrow(iris), 25, replace = TRUE)
iris_with_missing[idx_miss_factor, 5] <- NA
idx_miss_numeric <- sample(nrow(iris), 25, replace = TRUE)
iris_with_missing[idx_miss_numeric, 3] <- NA

x <- iris_with_missing[,-1]
y <- iris_with_missing[, 1]

forest <- forestry(x, y, ntree = 500, seed = 2,nthread = 2)
imputed_x <- impute_features(forest, x, seed = 2)