@kanaries/ml API Reference

This catalog describes every module available in @kanaries/ml so you can quickly locate the algorithm or utility that fits your JavaScript or TypeScript project. Each section below links to a detailed page covering parameters, usage patterns, and practical tips.

• Clusters – Implement k-means, DBSCAN, spectral clustering, and other unsupervised techniques for segmentation and feature discovery.
• Decomposition – Run dimensionality reduction methods such as PCA and SVD directly in the browser or Node.js.
• Ensemble – Combine multiple estimators including Isolation Forest and AdaBoost for robust anomaly detection and boosting workflows.
• Linear – Access regression and classification algorithms like Linear Regression, Lasso, and Logistic Regression with sklearn-like APIs.
• Manifold – Visualize high-dimensional data using t-SNE, Isomap, and related manifold learning approaches.
• Neighbors – Use k-nearest neighbors for search, recommendation, and classification tasks.
• SVM – Train support vector machines for high-margin classification and regression scenarios.
• Tree – Build decision trees and random forests for interpretable models and tabular data analysis.
• Bayes – Apply naive Bayes and probabilistic models to text classification, spam detection, and more.
• NeuralNetwork – Prototype lightweight neural networks tailored for in-browser inference.
• Utils – Leverage preprocessing helpers, metrics, and shared utilities to streamline your ML pipelines.
• SemiSupervised – Combine labeled and unlabeled data to boost performance in low-label environments.

Looking for inspiration? The Getting Started guide and examples directory provide end-to-end workflows you can adapt for your application.