This repository provides code examples and resources for the Boostcamp AI Tech Product Serving course, focusing on building and deploying machine learning models. It's designed for AI engineers and students learning practical MLOps skills, offering hands-on experience with batch and online serving, Dockerization, and model management.

How It Works

The project is structured into modules covering key MLOps components: batch serving with Airflow, online serving with FastAPI, Docker for containerization, and model management using MLflow. This modular approach allows learners to build a comprehensive understanding of the ML serving pipeline. The emphasis is on hands-on coding, encouraging users to type code directly, debug errors, and explore improvements.

Quick Start & Requirements

• Install: Follow instructions within each module's directory.
• Prerequisites: Python 3.x, Docker, Airflow, FastAPI, MLflow. Specific versions and additional libraries are detailed in individual module documentation.
• Resources: Requires local or cloud environment for running Airflow, FastAPI, and MLflow.

Highlighted Details

• Batch serving implementation using Apache Airflow.
• Online serving demonstration with FastAPI.
• Dockerization for containerizing serving applications.
• Model lifecycle management with MLflow.

Maintenance & Community

This project is associated with the Boostcamp AI Tech program. Community support and discussions are available via a dedicated Slack channel. The project adheres to Conventional Commits for contributions.

Licensing & Compatibility

The repository's license is not explicitly stated in the provided README. Compatibility for commercial use or closed-source linking is not specified.

Limitations & Caveats

This repository contains course materials and code examples, not a production-ready framework. Users are expected to adapt and extend the provided code for their specific needs. The project's primary focus is educational, and it may not cover all aspects of robust production serving.