Development build. This page describes main, not a released package. The latest published Lodestar.Preprocessing is 0.1.0 — read its documentation.

Home › Preprocessing › Feature scaling

MinMaxScalerOptions

The range MinMaxScaler maps each feature onto, and whether it clips to it.

public sealed record MinMaxScalerOptions

Properties — Low and High are the bounds each feature's minimum and maximum map to, defaulting to 0 and 1 as sklearn.preprocessing.MinMaxScaler's feature_range does. Clip bounds Transform's output to them, off by default.

Example — a range that is not the unit interval.

using Lodestar.Preprocessing;

double[] samples = [0.0, 2.0, 4.0];

MinMaxScaler scaler = MinMaxScaler.Fit(
    samples, featureCount: 1, new MinMaxScalerOptions { Low = -5.0, High = 3.0 });

double bottom = scaler.Transform(samples)[0];  // => -5
double top = scaler.Transform(samples)[2];     // => 3

Remarks — the two bounds are separate properties rather than a tuple, because a record with named members is what a caller reads back; feature_range is scikit-learn's spelling of the same pair. A low bound at or above the high one is refused, as is a non-finite one.

Applies to — net10.0, netstandard2.0.

See also — MinMaxScaler, MinMaxScaler.Fit.