- 🐎 Recovery-time prediction pipeline for endurance horse racing — stacked LightGBM + GBT, MAE within ±1.8–3.8 min
- 🤖 Multi-agent RAG — LangGraph + ChromaDB + Ollama, fully local
- 🔥 Fire & smoke detection — computer vision with RetinaNet (ResNet-50 + FPN)
- 🛒 CLV prediction on 541K real transactions, built end-to-end
- ❤️ Heart disease classifier — XGBoost + DNN, AUC ~0.80, SHAP, deployed on Streamlit
Highlights
- Pro
Pinned Loading
-
Muster
Muster PublicA web-based platform designed to centralize and visualize academic data (attendance, grades, assignments) with dashboards for professors, students, and parents. with AI-driven insights including pr…
Blade
-
Heart-Disease-Risk-Prediction-Patient-Clustering-
Heart-Disease-Risk-Prediction-Patient-Clustering- PublicHeart disease risk prediction & patient clustering using 70,000 records. Full ML pipeline: data preprocessing, EDA, KMeans/DBSCAN clustering, XGBoost & Random Forest classifiers, DNN + Autoencoder …
Jupyter Notebook 1
-
Customer-Value-Prediction-Segmentation
Customer-Value-Prediction-Segmentation PublicThis project analyzes the Online Retail dataset to uncover customer behavior and sales patterns. It includes CLV and RFM analysis, product and country performance, time-series trends, cohort retent…
Jupyter Notebook
-
Rag-egyptian-schools
Rag-egyptian-schools PublicMulti-agent RAG system for querying Egyptian private universities using LangGraph, Chroma, and Streamlit
Python
-
Music-Popularity-Prediction-Audio-Pattern-Discovery
Music-Popularity-Prediction-Audio-Pattern-Discovery PublicMusic popularity prediction and audio pattern discovery using 114,000 Spotify tracks. Full ML pipeline: data preprocessing, EDA, KMeans/DBSCAN/hierarchical clustering, Random Forest and XGBoost reg…
Jupyter Notebook
If the problem persists, check the GitHub status page or contact support.

