Learn R
mlr (version 2.10)

imputations: Built-in imputation methods.

Description

The built-ins are:
  • imputeConstant(const) for imputation using a constant value,
  • imputeMedian() for imputation using the median,
  • imputeMode() for imputation using the mode,
  • imputeMin(multiplier) for imputing constant values shifted below the minimum using min(x) - multiplier * diff(range(x)),
  • imputeMax(multiplier) for imputing constant values shifted above the maximum using max(x) + multiplier * diff(range(x)),
  • imputeNormal(mean, sd) for imputation using normally distributed random values. Mean and standard deviation will be calculated from the data if not provided.
  • imputeHist(breaks, use.mids) for imputation using random values with probabilities calculated using table or hist.
  • imputeLearner(learner, features = NULL) for imputations using the response of a classification or regression learner.

Usage

imputeConstant(const)
imputeMedian()
imputeMean()
imputeMode()
imputeMin(multiplier = 1)
imputeMax(multiplier = 1)
imputeUniform(min = NA_real_, max = NA_real_)
imputeNormal(mu = NA_real_, sd = NA_real_)
imputeHist(breaks, use.mids = TRUE)
imputeLearner(learner, features = NULL)

Arguments

const
[any] Constant valued use for imputation.
multiplier
[numeric(1)] Value that stored minimum or maximum is multiplied with when imputation is done.
min
[numeric(1)] Lower bound for uniform distribution. If NA (default), it will be estimated from the data.
max
[numeric(1)] Upper bound for uniform distribution. If NA (default), it will be estimated from the data.
mu
[numeric(1)] Mean of normal distribution. If missing it will be estimated from the data.
sd
[numeric(1)] Standard deviation of normal distribution. If missing it will be estimated from the data.
breaks
[numeric(1)] Number of breaks to use in hist. If missing, defaults to auto-detection via “Sturges”.
use.mids
[logical(1)] If x is numeric and a histogram is used, impute with bin mids (default) or instead draw uniformly distributed samples within bin range.
learner
[Learner | character(1)] Supervised learner. Its predictions will be used for imputations. If you pass a string the learner will be created via makeLearner. Note that the target column is not available for this operation.
features
[character] Features to use in learner for prediction. Default is NULL which uses all available features except the target column of the original task.

See Also

Other impute: impute, makeImputeMethod, makeImputeWrapper, reimpute