Univariate imputer for completing missing values with simple strategies.

Options

• :missing_values - The placeholder for the missing values. All occurrences of :missing_values will be imputed. The default value is :nan.

• :strategy - The imputation strategy.

  • :mean - replace missing values using the mean along each column.

  • :median - replace missing values using the median along each column.

  • :mode - replace missing using the most frequent value along each column. If there is more than one such value, only the smallest is returned.

  • :constant - replace missing values with :fill_value.

  The default value is :mean.

• :fill_value - When strategy is set to :constant, :fill_value is used to replace all occurrences of :missing_values. The default value is 0.0.

Return Values

The function returns a struct with the following parameters:

• :missing_values - the same value as in :missing_values

• :statistics - The imputation fill value for each feature. Computing statistics can result in Nx.Constant.nan/0 values.

Examples

iex> x = Nx.tensor([[1, 2, :nan], [3, 7, :nan], [:nan, 4, 5]])
iex> Scholar.Impute.SimpleImputer.fit(x, strategy: :mean)
%Scholar.Impute.SimpleImputer{
  statistics: Nx.tensor(
    [2.0, 4.333333492279053, 5.0]
  ),
  missing_values: :nan
}

Impute all missing values in x using fitted imputer.

Return Values

The function returns input tensor with NaN replaced with values saved in fitted imputer.

Examples

iex> x = Nx.tensor([[1, 2, :nan], [3, 7, :nan], [:nan, 4, 5]])
iex> imputer = Scholar.Impute.SimpleImputer.fit(x, strategy: :mean)
iex> Scholar.Impute.SimpleImputer.transform(imputer, x)
Nx.tensor(
  [
    [1.0, 2.0, 5.0],
    [3.0, 7.0, 5.0],
    [2.0, 4.0, 5.0]
  ]
)

iex> x = Nx.tensor([[1, 2, :nan], [3, 7, :nan], [:nan, 4, 5]])
iex> y = Nx.tensor([[7, :nan, 6], [6, 9, :nan], [8, :nan, 1]])
iex> imputer = Scholar.Impute.SimpleImputer.fit(x, strategy: :median)
iex> Scholar.Impute.SimpleImputer.transform(imputer, y)
Nx.tensor(
  [
    [7.0, 4.0, 6.0],
    [6.0, 9.0, 5.0],
    [8.0, 4.0, 1.0]
  ]
)