Pure Zig implementation of Jinja2 templating. Bytecode compilation, 40+ filters, template inheritance, macros, autoescaping, and sandboxing. Built for HuggingFace ML pipelines and high-performance applications.

A LOWL-to-LLVM map for compiling the ML/I Macro Processor.

This repository explores AI/ML approaches towards automated tracking of dietary intake: estimating the ingredients & macro-nutrients in a food plate using digital images.

Advanced multi-factor market downturn prediction model using options data, volatility metrics, gamma exposure, jump risk, and macro sentiment. Features real-time data integration, regime detection, ensemble ML, risk-adjusted signals, and backtesting for precise early warning of market stress events.

This project highlights a Spark application built on Scala. It utilizes Spark Core, Spark SQL and Spark ML (Machine Learning libraries) for predicting stock prices of specific airline companies. We have used the Google trending words (searched on internet and relevant to financial domain) and also macro-economic oil prices as alternate data to predict stock prices.

Event-driven ML end-to-end pipeline for small-cap biotech (news/NLP, options-implied metrics, macro/tech). Purged k-fold + embargo, regime gate, deterministic backtester.

FE800 Financial Engineering project on market regime detection and regime-aware portfolio allocation. Compares Hidden Markov, Semi-Markov, Jump Models, RHSM, and ML models using macro, volatility, and market indicators to dynamically allocate across equities, bonds, and alternatives.

Collection of helper macros for machine learning in Rust

Test cases of the features of the ancient macro processor ML/I

DBT macros and tests for advanced statistical analysis, ML refactors from sk-learn and other libraries, and scalable big data workflows

SAS macro "ml" implements maximum-likelihood estimation via PROC NLMIXED as described in the paper:

🥇 Analyse & aide à la décision pour métaux précieux — ML walk-forward (XGBoost, LSTM), backtesting, macro. Streamlit PWA.

A two-stage ML framework combining return prediction and macro-driven gating to improve trading decisions under noisy financial signals.

A smart health app that recommends personalized diet and water intake plans based on user data and goals. Tracks meals, adjusts macros and hydration dynamically, and uses ML to suggest nutrient-rich meals tailored for weight loss, muscle gain, or maintenance.

AI-powered stock forecasting dashboard built with Prophet ML, live yFinance & FRED macro data, risk metrics (Sharpe, Sortino, Max Drawdown) and interactive Plotly charts — deployed on Streamlit

A Python-based platform simulating 5G/6G networks with ML integration. Features beamforming, MIMO optimization, terahertz communication, and edge computing. Uses ML for traffic prediction, QoS analysis, and anomaly detection. Processes real-time metrics across macro/small cells with interactive visualization.

Advanced DSL for economic modeling with AI-powered agents, automatic equilibrium solving, causal analysis, and unified DSGE/ABM simulation. Features ML integration, policy experiments, and macro-micro bridge engine.

A React Native app for managing type 2 diabetes or pre-diabetes with the help of machine learning. Featuring glucose logging, ML-driven risk alerts, macro recommendations and trend data visualizations.

This notebook is the work of Varun Suryan in response to the application for ML Scientist role at Macro-Eyes.The two parts of this assignment are answered in this notebook sequentially.

Reproducible classical ML baselines for class‑imbalanced intrusion detection on UNSW‑NB15 (binary + multiclass) comparing no balancing, class weighting, and SMOTE with macro‑F1, ROC‑AUC, G‑Mean, and per‑class metrics.

An end-to-end autonomous quantitative research and execution pipeline. This system synchronizes live market data, macro-economic indicators, and an ensemble ML architecture to detect and forecast market regimes (Bearish, Neutral, Bullish).

Python-based regime detection engine for factor timing and portfolio allocation. Combines macro indicators, market data, and ML classification to identify global risk regimes and improve Sharpe through adaptive weighting and config-driven backtesting.

Proyek ini membangun model ML untuk memprediksi tipe penghargaan misi dalam kompetisi eksplorasi antarbintang AIRena Corp. Model dikembangkan menggunakan berbagai dataset sebagai tugas klasifikasi multi-kelas. Evaluasi dilakukan dengan F1 Score (Macro Average). Projek Untuk Lomba AIrena 2025

Conducted Twitter sentiment analysis on Google and Apple products. Employed a range of supervised ML algorithms and neural networks to tackle an imbalanced NLP multiclass classification problem. Selected the model with the highest macro F1 score for evaluation.

Forecast next-month S&P 500 equity premium using classic macro/valuation predictors and compare econometric vs. ML models under consistent rolling/expanding backtests; evaluate with OOS 𝑅 2 R 2 and trading Sharpe; includes an IBM single-stock extension.

Amazon Product Quality Predictor Developed a weakly supervised ML classifier for Amazon products using proxy labeling from text richness and price anomaly signals. Built TF-IDF features and trained a Logistic Regression model on ~75K products, achieving Macro F1 ≈ 0.82, with a FastAPI dashboard for real-time analysis.

Macros for D2H ML local analysis

No description available

collection of macros used for processing simulation files

No description available