Interface MaxAbsScaler

public interface MaxAbsScaler
Scales each feature by its maximum absolute value. This class scales and translates each feature individually such that the maximal absolute value of each feature in the training set will be 1.0. It does not shift/center the data, and thus does not destroy any sparsity.

The maximum absolute value accumulator is initialized at 0.0, so even all-negative columns produce the correct result (e.g., [-3,-2,-1] yields scale = 3.0). When all values are zero the scale falls back to 1.0.

  • Method Details

    • fit

      static InvertibleColumnTransform fit(DataFrame data, String... columns)
      Fits the data transformation.
      Parameters:
      data - the training data.
      columns - the columns to transform. If empty, transform all the numeric columns.
      Returns:
      the transform.
      Throws:
      IllegalArgumentException - if the data frame is empty or a specified column is non-numeric.