@kanaries/ml

@kanaries/ml is a JavaScript/TypeScript library that provides a set of machine learning algorithms with an API similar to scikit-learn. It works in both browsers and Node.js environments.

Features

• Classification and regression trees
• k-nearest neighbors
• Support vector machines
• Naive Bayes classifier
• Clustering algorithms (KMeans, DBSCAN, OPTICS, Mean Shift, HDBSCAN)
• Dimensionality reduction (PCA)
• Manifold learning (t-SNE)
• Basic linear algebra utilities

Installation

# using yarn
yarn add @kanaries/ml

# or npm
npm install @kanaries/ml

Quick Start

import { Neighbors } from '@kanaries/ml';

const trainX = [
    [0.12, 0.2, /* ... */ 0.2],
    [0.21, 0.3, /* ... */ 0.2],
];
const trainY = [0, 1];

const knn = new Neighbors.KNearstNeighbors(3, 'distance', '2-norm');
knn.fit(trainX, trainY);

const testX = [
    [0.52, 0.72, /* ... */ 0.24],
    [0.11, 0.98, /* ... */ 0.32],
];
const result = knn.predict(testX);
console.log(result);

Supported Algorithms

The library exposes several categories of algorithms:

• Tree: DecisionTreeClassifier, DecisionTreeRegressor, ExtraTreeClassifier, ExtraTreeRegressor
• Neighbors: KNearstNeighbors, BallTree, KDTree
• Linear Models: LinearRegression, LogisticRegression
• Support Vector Machines: SVC, NuSVC, LinearSVC
• Naive Bayes: BernoulliNB, CategoricalNB
• Clustering: KMeans, kmeansPlusPlus, DBScan, OPTICS, MeanShift, HDBScan
• Decomposition: PCA
• Manifold Learning: SpectralEmbedding, MDS, LocallyLinearEmbedding, TSNE
• Ensemble: IsolationForest, AdaBoostClassifier
• Utilities: linear algebra helpers and math functions

Advanced Features

asyncMode

asyncMode runs a synchronous function in a worker (Web Worker or Node.js worker thread) and returns a Promise.

import { utils } from '@kanaries/ml';

const heavy = (x: number) => x * x;
const runAsync = utils.asyncMode(heavy);

const result = await runAsync(5);

Development

# Install dependencies
yarn

# Run tests
npm run test

# Build the library
yarn build

# Start the example development server
yarn dev