Overview
MinMaxScaler transforms features by scaling each feature to a given range. The default range is [0, 1].
Constructor
new MinMaxScaler(options?: MinMaxScalerOptions)
options.featureRange
[number, number]
default:"[0, 1]"
Desired range of transformed data as [min, max]. Both values must be finite and min must be less than max.
Properties
Per feature minimum seen in the training data.
Per feature maximum seen in the training data.
Per feature range (dataMax_ - dataMin_) in the training data.
Per feature scaling factor.
Per feature minimum offset.
Methods
fit
Compute the minimum and maximum to be used for later scaling.
Training data matrix where each row is a sample and each column is a feature.
this - The fitted scaler instance.
transform(X: Matrix): Matrix
Data matrix to transform.
Matrix - Transformed data scaled to the specified range.
fitTransform(X: Matrix): Matrix
fit(X).transform(X).
Training data matrix to fit and transform.
Matrix - Transformed data.
inverseTransform(X: Matrix): Matrix
Transformed data matrix to inverse transform.
Matrix - Original scale data.
Example
import { MinMaxScaler } from 'bun-scikit';
// Training data
const X = [
[1, 2],
[2, 4],
[3, 6],
[4, 8]
];
// Scale to [0, 1] (default)
const scaler = new MinMaxScaler();
const XScaled = scaler.fitTransform(X);
console.log('Scaled [0,1]:', XScaled);
// Output: Scaled [0,1]: [[0, 0], [0.333, 0.333], [0.667, 0.667], [1, 1]]
// Scale to custom range [-1, 1]
const scaler2 = new MinMaxScaler({ featureRange: [-1, 1] });
const XScaled2 = scaler2.fitTransform(X);
console.log('Scaled [-1,1]:', XScaled2);
// Output: Scaled [-1,1]: [[-1, -1], [-0.333, -0.333], [0.333, 0.333], [1, 1]]
// Check fitted parameters
console.log('Data min:', scaler.dataMin_);
// Output: Data min: [1, 2]
console.log('Data max:', scaler.dataMax_);
// Output: Data max: [4, 8]
// Inverse transform
const XOriginal = scaler.inverseTransform(XScaled);
console.log('Original:', XOriginal);
// Output: Original: [[1, 2], [2, 4], [3, 6], [4, 8]]
Notes
- Features with zero range (constant values) are not scaled
- The scaler must be fitted before calling
transform() or inverseTransform()
- Input data must be finite (no NaN or Infinity values)
- Useful when features need to be in a specific bounded range