MLOps Challenges

This repository contains hands-on challenges for end-to-end machine learning operations (MLOps) with Azure Machine Learning.

To complete these exercises, you’ll need a Microsoft Azure subscription. If your instructor has not provided you with one, you can sign up for a free trial at https://azure.microsoft.com.

Labs

Find the best classification model with Azure Machine Learning

Experiment with automated machine learning and interactive notebooks to train machine learning models.

Optimize model training in Azure Machine Learning

Optimize model training with scripts and track with MLflow.

Perform hyperparameter tuning with a sweep job

Perform hyperparameter tuning with a sweep job.

Run pipelines in Azure Machine Learning

Run pipelines in Azure Machine Learning.

Plan and prepare an MLOps solution with Azure Machine Learning

Design dev and prod environments and plan Azure CLI automation to provision workspaces, registries, and data assets.

Automate model training with GitHub Actions

Securely integrate GitHub with Azure Machine Learning and automate model training with GitHub Actions workflows.

Deploy and monitor a model in Azure Machine Learning

Use GitHub Actions pull requests, environments, and comment triggers to train models in dev and prod, deploy to a managed online endpoint, and configure Azure Machine Learning model monitoring.