MLOps Notes

Why MLOps

Quicker experimentation cycle -> More models

Quicker productization cycle -> More models in production

Full traceability for all models -> More models in production safely and scalability"

Tools for MLOps

• Data Analysis - Python, Pandas
• Source Control  - Git
• Test & Build Services - PyTest & Make
• Deployment Services - Git, DVC
• Model & Dataset Registry - DVC[aws s3]
• Feature Store - Project code library
• ML Metadata Store - DVC
• ML Pipeline Orchestrator - DVC & Make
• Experimentation Tracking - MLFlow

Mops Stack

GCP MLOPs

• MLOps
• Featurestore
• Model Monitoring

AWS MLOps

• Feature store
• Model Monitoring

Azure MLOps

• Feature store
• Model Monitoring
• Set up MLOps

MLFlow

MLflow - Tracking experiments, Packaging ML code, Managing and deploying models, central model store  

Made with ML

• Made With ML
• CI/CD for Machine Learning
• Model Serving
• Distributed training
• Monitoring Machine Learning Systems

MLOps Courses

• MLOps Zoomcamp
• MLOps-Basics
• MLOps Guide
• awesome-mlops
• dtu_mlops
• MLOps Course
• MLOps Tutorial #1: Intro to Continuous Integration for ML

Kubernetes

• Intro to Kubernetes Workshop
• Kubernetes slides
• Kubernetes The Hard Way
• From Notebook to Pipeline in No Time with LineaPy and DVC - Thomas Fraunholz at PyData 

Keep Exploring!!!