Machine Learning Operations (MLOps) Challenge

Machine learning operations (MLOps) applies DevOps principles to machine learning projects. You'll learn how to implement key concepts like source control, automation, and CI/CD to build an end-to-end MLOps solution while using Python to train, save, and use a machine learning model.

Microsoft Badge: Get started with AI on Azure was issued by Microsoft

Microsoft Badge: Introduction to GitHub Copilot was issued by Microsoft

References

Microsoft Badge: Introduction to DevOps principles for machine learning was issued by Microsoft

Overview of resources

Microsoft Badge: Explore Azure Machine Learning workspace resources and assets was issued by Microsoft

https://ml.azure.com/

Microsoft Badge: Explore developer tools for workspace interaction was issued by Microsoft

Microsoft Badge: Design a data ingestion strategy for machine learning projects was issued by Microsoft

Microsoft Badge: Design a machine learning model training solution was issued by Microsoft

Microsoft Badge: Design a model deployment solution was issued by Microsoft

Microsoft Badge: Design a machine learning operations solution was issued by Microsoft

Microsoft Badge: Use an Azure Machine Learning job for automation was issued by Microsoft

Useful resources

Microsoft Badge: Trigger Azure Machine Learning jobs with GitHub Actions was issued by Microsoft

Useful resources

The introduction to DevOps principles for machine learning module covers how to integrate Azure Machine Learning with DevOps tools.
Use GitHub Actions with Azure Machine Learning.
Learn more about service principal objects in Azure Active Directory.
Learn more about encrypted secrets in GitHub, like how to name and how to create a secret in a GitHub repo.
Manually running a workflow in GitHub Actions.
Re-running workflows and jobs in GitHub Actions.
General documentation for GitHub Actions.

Microsoft Badge: Trigger GitHub Actions with trunk-based development was issued by Microsoft

Useful resources

Learn more about source control for machine learning projects and how to work with feature-based development and GitHub repos.
General documentation for GitHub Actions.
Triggering a GitHub Actions workflow.
Events that trigger workflows.
Workflow syntax for GitHub Actions.

Microsoft Badge: Work with linting and unit testing in GitHub Actions was issued by Microsoft

Useful resources

Flake8 documentation, including error codes and their descriptions.
A beginner’s guide to Python testing.
Learn more about test infrastructure using Azure ML and how to create tests.
Learn more about testing with Pytest.

In this challenge, all testing is executed with GitHub Actions. Optionally, you can learn how to verify your code locally with Visual Studio Code.

Tip

Learn more about how to work with source control for machine learning projects, including trunk-based development and verifying your code locally.

Learn more about how to run unit tests with Pytest.

You can verify code automatically with GitHub Actions, or manually in Visual Studio Code. Learn more about how to verify your code locally.

Microsoft Badge: Work with environments in GitHub Actions was issued by Microsoft

integrate Azure Machine Learning with DevOps tools such as GitHub

Tip

Learn more about how to use environments in GitHub Actions and how to add approvals.

Useful resources

Tip

Learn more about how to deploy MLflow models.

Learn more about how to deploy a model with the Azure Machine Learning CLI (v2).

Useful resources

Microsoft Trophy: End-to-end machine learning operations (MLOps) with Azure Machine Learning was issued by Microsoft

learn.microsoft.com/en-us/training/achievements/learn.wwl.build-first-machine-operations-mlops-workflow.trophy?username=DBVargas&sharingId=8A823BAF779A77B6

Report abuse