Crate kategorize 

K-modes and K-prototypes Clustering

This crate provides implementations of the k-modes and k-prototypes clustering algorithms for categorical and mixed (categorical + numerical) data.

Features

• K-modes: Clustering for purely categorical data
• K-prototypes: Clustering for mixed categorical and numerical data
• Multiple initialization strategies (Huang, Cao, random)
• Parallel processing support via Rayon
• Comprehensive distance metrics for categorical data

Example

use kategorize::{KModes, InitMethod};
use ndarray::Array2;

// Create sample categorical data
let data = Array2::from_shape_vec((4, 2), vec![
    "A", "X", "A", "Y", "B", "X", "B", "Y"
]).unwrap();

// Create k-modes clusterer
let mut kmodes = KModes::new(2)
    .init_method(InitMethod::Huang)
    .max_iter(100)
    .n_init(10);

// Fit the model and get cluster assignments
let result = kmodes.fit(data.view()).unwrap();
println!("Cluster labels: {:?}", result.labels);

Re-exports

pub use error::Error;

pub use error::Result;

pub use kmodes::KModes;

pub use kmodes::KModesResult;

pub use kprototypes::KPrototypes;

pub use kprototypes::KPrototypesResult;

pub use kprototypes::MixedValue;

pub use initialization::InitMethod;

Modules

distance

Distance metrics for categorical and mixed data

error

Error types for the k-modes crate

initialization

Initialization methods for k-modes and k-prototypes clustering

kmodes

K-modes clustering algorithm implementation

kprototypes

K-prototypes clustering algorithm for mixed categorical and numerical data

utils

Utility functions for k-modes clustering

Type Aliases

Array1

Re-export commonly used types from ndarray one-dimensional array

Array2

Re-export commonly used types from ndarray two-dimensional array

ArrayView1

Re-export commonly used types from ndarray one-dimensional array view

ArrayView2

Re-export commonly used types from ndarray two-dimensional array view