ClusterKit provides state-of-the-art dimensionality reduction and clustering algorithms accessible inside Ruby through native Rust bindings. It supports UMAP, PCA, K-means, and HDBSCAN and runs within the Ruby process without requiring Python. Native Rust performance via Magnus FFI yields roughly 2-3x speedups with parallel processing. Seed support enables reproducible results. Robust handling of extreme data ranges prevents crashes by fixing the box_size panic. The library interoperates with Ruby arrays and Numo::NArray. Use cases include data visualization, density-based clustering, anomaly detection, and exploratory data analysis without leaving Ruby.
ClusterKit brings state-of-the-art dimensionality reduction and clustering algorithms to Ruby through native Rust bindings. Run UMAP, PCA, K-means, and HDBSCAN directly in your Ruby process with performance matching Python's scikit-learn - no Python required. What makes it special: Native Rust performance via Magnus FFI (2-3x faster with parallel processing) Reproducible results with seed support Handles extreme data ranges without crashing (fixed the notorious box_size panic) Works seamlessly with Ruby arrays and Numo::NArray
process with performance matching Python's scikit-learn - no Python required. What makes it special: Native Rust performance via Magnus FFI (2-3x faster with parallel processing) Reproducible results with seed support Handles extreme data ranges without crashing (fixed the notorious box_size panic) Works seamlessly with Ruby arrays and Numo::NArray Perfect for data visualization, density-based clustering, anomaly detection, and exploratory data analysis - all without leaving Ruby or shelling out to Python.
Collection
[
|
...
]