Draw samples from a multi-variate gaussian distribution

Machine learning tools

Fast multidimensional filtering for coordinated views.

[Hypertune](https://www.hypertune.com/) is the most flexible platform for feature flags, A/B testing, analytics, and app configuration. Built with full end-to-end type safety, Git-style version control and local, synchronous, in-memory flag evaluation. Op

No description provided.

Construct AI & ML models with JSON using Typescript & Tensorflow

Kalman filter (and Extended Kalman Filter) Multi-dimensional implementation in Javascript

StreetJS feature-flag foundation: typed boolean/multivariate flags with attribute targeting rules and deterministic percentage rollouts (stable per-subject bucketing), a pluggable flag store, and a fast in-memory registry. Zero runtime dependencies; edge/

Rolling ball baseline correction

Direct - DIviding RECTangles optimization algorithm

XMBL identity and signatures — MAYO post-quantum signing plus the experimental cube-curve crypto seam.

Scaffold, deploy, and manage your Flaghoist feature-flag service from the terminal.

GeoJSON morphing utilities with MapLibre-first adapter and Leaflet compatibility

LaunchFlag feature flags client. One key, one function, edge-safe, fails safe.

A Model Context Protocol server for generating charts using QuickChart.io

Flag schema, evaluation engine, and storage/auth interfaces for Flaghoist. Zero dependencies.

Cheminformatics support: import, rendering, sketching, calculation of properties, predictive models, augmentations, multiple analyses.

High-performance LOWESS smoothing for Node.js

Feature flags & A/B experiments with no SaaS and no infrastructure — deterministic bucketing (same user always gets the same result), targeting rules, percentage rollouts and weighted variants. Synchronous, zero-dependency, isomorphic.

High-performance LOWESS (Locally Weighted Scatterplot Smoothing)