river-ml

Complete system for the classification of Activities of Daily Living (ADL) by collecting inertial data from smartphones and evaluating supervised models (RF, SVM) under a Stream Learning approach, including online architecture for real-time classification.

Python framework for online learning in machine learning with streaming data pipelines, concept drift detection, and prequential evaluation.

A 100% locally-run Host SOC and Intrusion Detection System (IDS) for Ubuntu Linux featuring real-time journalctl streaming, online anomaly detection with River ML, local SQLite persistence, and an interactive Rich CLI dashboard.

Client-server backend for real-time Activity of Daily Living (ADL) classification. Handles window ingestion, prediction, label requests, and incremental model updates.

Real-time NYC subway anomaly detection: GTFS-RT, online ML (River), live Mapbox command center tracking 1000+ stations.

Live ML-powered trading dashboard for NinjaTrader 8 — 10 incremental models with distinct trading personalities that learn in real time, compete via Champion/Challenger, and level up based on simulated P&L.