Overview
Imbalanced-learn is a Python package offering a number of re-sampling techniques commonly used in datasets showing strong between-class imbalance. It is compatible with scikit-learn and is part of the scikit-learn-contrib projects.
Why Imbalanced-learn Matters
In real-world datasets, it’s common to have imbalanced classes where one class significantly outnumbers others. This creates challenges:
- Credit Card Fraud: 99.9% legitimate transactions vs 0.1% fraudulent
- Medical Diagnosis: Rare diseases with few positive cases
- Manufacturing Defects: Most products pass quality control
- Customer Churn: Typically only 2-5% of customers churn
Standard machine learning algorithms often fail on imbalanced datasets, predicting only the majority class.