[0.1.0] - 2025-06-14

Added

• Initial release of PyIVM (Python Library for Clustering Quality Metrics)
• Implementation of six essential clustering validation metrics:
  • Calinski-Harabasz Index
  • Davies-Bouldin Index
  • Dunn Index
  • I-Index
  • Silhouette Coefficient
  • Xie-Beni Index
• Support for both original and adjusted forms of all metrics
• Adjusted metrics provide bias-free evaluation and consistent "higher = better" interpretation
• Simple API compatible with scikit-learn and numpy arrays
• Comprehensive test suite with sanity checks
• Support for Python 3.9+
• Dependencies: NumPy, SciPy, scikit-learn, pandas
• Complete documentation with usage examples
• Poetry-based project structure for easy development
• MIT License

Features

• All metrics follow consistent pyivm.metric_name(X, labels, adjusted=False) API
• Adjusted metrics enable fair comparison across different numbers of clusters
• Simple installation via pip install pyivm
• Comprehensive README with quick start guide and API reference

Technical Details

• Based on research published in IEEE TPAMI 2025
• Implements theoretical foundations for adjusted clustering validation metrics
• Optimized for performance with NumPy and SciPy backends