MLOps Class Project – Reproducible ML Pipelines

This repository is part of my MLOps coursework in the Master of Science in Data Science program at the University of San Francisco. The goal of the class is to learn how to build reliable, reproducible, and production-ready machine learning workflows.

---

🎯 Project Goals

• Develop an end-to-end ML pipeline from data ingestion to model deployment
• Practice building modular code and workflows using MLOps tools
• Learn how to track data, parameters, and experiments
• Collaborate using Git and GitHub best practices

---

🛠️ Tools and Technologies

Tool	Purpose
Python	Main programming language
DVC	Data & model version control
Git / GitHub	Code versioning and collaboration
VS Code	Dev environment
Pandas, scikit-learn	Data wrangling & preprocessing

---

📦 Current Milestone

✅ Milestone 1: Data Preprocessing + DVC Integration

• Clean and encode data
• Track raw and processed datasets with DVC
• Reproducible preprocessing script with dvc.yaml

Next steps:

• 🧠 Add model training stage
• 📊 Add evaluation + metrics logging
• 🚀 Set up model serving (Flask or FastAPI)