Mubarak Adisa

GEMSTONE PRICE PREDICTION Gemstone (Diamond) Price Prediction – End-to-End Machine Learning System 🧠 Project Description Developed a full end-to-end machine learning system to predict gemstone (diamond) prices based on physical and quality attributes. This project covers the entire ML lifecycle, from data ingestion and exploratory data analysis to model training, deployment, and CI/CD automation. The solution is production-ready, containerized with Docker, and deployed to cloud platforms, demonstrating real-world MLOps practices. ⚙️ Key Contributions • Performed Exploratory Data Analysis (EDA) and feature engineering • Built preprocessing pipelines using ColumnTransformer and encoders • Trained and compared multiple regression models: o Linear Regression o Ridge, Lasso, ElasticNet o Random Forest Regressor o AdaBoost o XGBoost Regressor • Evaluated models and selected the best-performing one • Built a Flask web application for price prediction • Implemented CI/CD pipelines using GitHub Actions • Containerized the application using Docker & Docker Compose • Deployed the system on AWS and Azure Tools & Technologies Python, Pandas, NumPy Scikit-learn, XGBoost Flask Docker, Docker Compose GitHub Actions (CI/CD) AWS, Azure 📊 Results & Impact • Accurate gemstone price prediction using optimized ML models • Production-grade ML system suitable for e-commerce and pricing platforms • Demonstrates strong skills in ML engineering and deployment 🌐 Web Application (Flask) The web app has three main pages: • Home Page 🏠 – Introduction & navigation • Form Page 📝 – Input diamond features • Result Page 📈 – Predicted diamond price 🔗 GitHub Repository: https://github.com/AdMub/Gemstone-Price-Prediction-End-to-EndProjects