📏 ml_scaler #

A lightweight and pure Dart library for feature scaling and normalization, including MinMaxScaler and StandardScaler. Ideal for machine learning preprocessing.

✨ Features #

• 🔢 MinMaxScaler: Scale features to a defined range (default: 0–1)
• 🧮 StandardScaler: Zero mean and unit variance (Z-score)
• 🔁 fit(), transform(), inverseTransform() APIs
• 🧠 Serializable models with toModel() and loadFromModel()
• ✅ Fully tested and documented

🚀 Installation #

Add to your pubspec.yaml:

dependencies:
  ml_scaler: ^0.1.0

Then run:

dart pub get

📦 Usage Example #

import 'package:ml_scaler/ml_scaler.dart';

void main() {
  final data = [
    [1.0, 2.0],
    [2.0, 4.0],
    [3.0, 6.0],
  ];

  // MinMaxScaler
  final minMax = MinMaxScaler();
  minMax.fit(data);
  final scaled = minMax.transform(data);
  print('MinMax scaled: $scaled');

  // StandardScaler
  final stdScaler = StandardScaler();
  stdScaler.fit(data);
  final standardized = stdScaler.transform(data);
  print('Standard scaled: $standardized');
}

✅ Output Example #

MinMax scaled: [[0.0, 0.0], [0.5, 0.5], [1.0, 1.0]]
Standard scaled: [[-1.0, -1.0], [0.0, 0.0], [1.0, 1.0]]

📂 Directory Structure #

lib/
  ├── scalers/
  │   ├── min_max_scaler.dart
  │   ├── standard_scaler.dart
  ├── utils/
  │   └── data_validator.dart
  ├── models/
  │   └── scaler_model.dart
  └── ml_scaler.dart

🛡 License #

MIT © Mehmet Çelik