Statistical Utilities
Compute simple statistics with the @kanaries/ml Stat JavaScript and TypeScript utilities for browser and Node.js machine learning helpers.
Algorithm overview
The Stat utilities provide small helpers used by models and examples: mode, mean, probability conversion, entropy, and Gini impurity.
JavaScript implementation
@kanaries/ml exposes these functions under utils.Stat, so lightweight statistics can stay in the same JavaScript runtime as preprocessing, training, and inference.
Quick start example
import { utils } from '@kanaries/ml';
const majority = utils.Stat.mode([1, 1, 2, 3]);
const avg = utils.Stat.mean([1, 2, 3]);
const entropy = utils.Stat.entropy([2, 2, 4]);Detailed API reference
utils.Stat.mode(data: number[]): number
utils.Stat.mean(data: number[]): number
utils.Stat.freqs2Probs(freqs: number[]): number[]
utils.Stat.entropy(data: number[]): number
utils.Stat.gini(data: number[]): numberentropy and gini treat the input array as frequencies and convert it to probabilities internally.
Model Selection Utilities
Run cross-validation, K-fold splits, grid search, and randomized search with the @kanaries/ml ModelSelection JavaScript implementation.
asyncMode
Learn what asyncMode does, when to use it, and how to run synchronous machine learning work in JavaScript without blocking browser or Node.js execution.