MLOps_Zoomcamp_Study

This is the repository of my study in MLOps Zoomcamp from DataTalksClub.

MLOps, short for Machine Learning Operations, is a set of practices and tools designed to deploy and maintain machine learning models in production reliably and efficiently. It combines machine learning (ML) with DevOps (Development Operations) to automate and streamline the process of developing, testing, and deploying ML models.

Key Aspects of MLOps

• Automation
• Collaboration
• Monitoring
• Versioning
• Scability

Here is binder link to check Jupyter Notebook.

Module 1: Introduction

• Environment Preparation
• Machine Learning
• MLOps Overview
• MLOps Maturity Model
• Link

Module 2: Experiment Tracking

• Mlflow
• Setting up Mlflow Server
• Logging Parameters, Metrics, and Artifacts
• Model Registry
• Link

Module 3: Orchestration

• MageAI
• Setting up MageAI
• Building pipelines
• Integrating MLFlow for experiment tracking
• Link

Module 4: Deployment

• Three ways of deployment
  • Web service
  • Streaming
  • Batch
• Docker
• Link