M

machine-learning

Official repository of the Ruhr university Neural Network energy representation (RuNNer).

Full stack IoT predictive maintenance platform with embedded firmware, sensor telemetry, FastAPI services, PostgreSQL storage, dashboards and anomaly detection.

Aston University DG1AID lab repository with AI and data science notes, Python notebooks, NumPy, Pandas, search algorithms and machine learning practice.

Finally a smart RSS reader which doesn't suck ass or your data.

RuNNerASE is a collection of packages for training, evaluating, and analyzing machine learning potentials with RuNNer, the Ruhr university Neural Network energy representation.

A production-grade machine learning system for classifying celestial objects — stars, galaxies, and quasars — from 500,000 photometric observations sourced from the Sloan Digital Sky Survey (SDSS DR18). The pipeline covers LOF-based outlier detection, SMOTEENN class balancing, and SelectKBest feature selection, with six classifier configurations benchmarked against each other. Random Forest achieved the highest accuracy at ~99.51%, while LightGBM was selected for deployment due to its faster inference, smaller footprint, and clean ONNX export path. The system is fully containerized with Docker, backed by a CI/CD pipeline, and served live via a Gradio interface on Hugging Face Spaces.

Self-hosted forecasting + prediction service. Five zero-shot time-series foundation models (Chronos-2, TimesFM 2.5, Moirai-2, Toto-1, Sundial) across six forecast types, plus nine supervised tabular ML backends (LightGBM, XGBoost, sklearn family) with calibrated / stacking / diversified meta-learners. Unified REST API + MCP server.

Quantum Synapse is an open-source, non-invasive brain–computer interface platform that enables mind-controlled interaction through wearable EEG technology. It combines real-time neural signal acquisition, signal processing, and machine learning to translate brain activity into digital input. Designed for privacy, modularity, and accessibility, the project provides a research-focused framework for exploring human–computer interaction without surgical implants or proprietary systems. https://roxanneardary.com/quantum-synapse/

TFM: sistema FOSS de control hands-free de cursor mediante mirada y gestos para personas con paralisis motora severa.

StrataAI is an open-source, modular AI system that analyzes global gemstone markets to generate transparent, data-driven fair market valuations. By aggregating auction data, distributor listings, certification records, and geological origin information, it builds a reproducible pricing model for precious stones using machine learning and structured market intelligence. https://roxanneardary.com/strataai/

A machine learning pipeline designed to analyze sleep tracking features and predict daily subjective sleep quality ratings.

A data mining project analyzing hate crime patterns in the United States from 2017 to 2025, using clustering, predictive modeling, and association rule mining.

Data Science / Machine Learning Pipeline component for training and deploying ML models using CI

A deterministic verification layer for AI systems. QWED verifies AI outputs using mathematics, symbolic reasoning, and formal methods (Z3, SMT, SymPy), creating an auditable trust boundary for agentic AI. Not generation. Verification.

PulseCheck is an open-source dynamic code intelligence platform that monitors and analyzes code at runtime. It detects TODOs, decoy or unused functions, anti-distillation tricks, and security/safety issues. Leveraging AI and machine learning, PulseCheck provides anomaly detection, predictive TODO analysis, behavior clustering, and runtime insights, giving developers actionable intelligence to secure, optimize, and complete their code from draft to deployed. https://roxanneardary.com/pulsecheck/

Lightweight vector embeddings store for RAG applications

Neural network experiments for Supercompress compression engine.

End-to-end predictive maintenance ML pipeline for hydraulic systems with Streamlit app, XGBoost models, SHAP, and real-world sensor data.

https://hydraulic-monitoring.streamlit.app/

About Behavioral finance meets machine learning. Early-warning system to forecast S&P 500 downturns using sentiment, volatility, and unemployment data — with SHAP explainability ,Gradio and Hugging Face deployment.

https://huggingface.co/spaces/Artur-Melnyk/Market-Mood-Forecasting

Explainable suburb similarity explorer built with Python, Streamlit, vector search and transparent model reasoning.