Crate anofox_ml_neighbors 

K-nearest neighbors classifiers and regressors with KD-tree acceleration.

This crate provides KnnClassifier and KnnRegressor, both of which support configurable distance metrics (DistanceMetric) and weighting strategies (WeightFunction). When Euclidean distance is selected, a KD-tree is built at fit time for efficient O(log n) neighbor lookups; other metrics use brute-force search.

Examples

use ndarray::array;
use anofox_ml_core::{Fit, Predict};
use anofox_ml_neighbors::KnnClassifier;

let x_train = array![
    [0.0, 0.0],
    [1.0, 1.0],
    [2.0, 2.0],
    [10.0, 10.0],
    [11.0, 11.0],
    [12.0, 12.0]
];
let y_train = array![0.0, 0.0, 0.0, 1.0, 1.0, 1.0];

let knn = KnnClassifier::new(3);
let fitted = Fit::fit(&knn, &x_train, &y_train).unwrap();

let x_test = array![[0.5, 0.5], [10.5, 10.5]];
let preds = fitted.predict(&x_test).unwrap();
assert!((preds[0] - 0.0_f64).abs() < 1e-10);
assert!((preds[1] - 1.0_f64).abs() < 1e-10);

Re-exports

pub use distance::DistanceMetric;

pub use knn_classifier::FittedKnnClassifier;

pub use knn_classifier::KnnClassifier;

pub use knn_classifier::WeightFunction;

pub use knn_regressor::FittedKnnRegressor;

pub use knn_regressor::KnnRegressor;

pub use lof::FittedLocalOutlierFactor;

pub use lof::LocalOutlierFactor;

pub use lof::LofAlgorithm;

Modules

distance

kdtree

knn_classifier

knn_regressor

lof

Local Outlier Factor.