Azure Machine Learning services for AZ-900

Today we will not go very deep in that topic. It is rather to show you a basic services for AI/ML. Currently Artificial Intelligence is a buzzword and when we think about it LLMs and GPTs come to out minds, but threre are more technical services, added to Azure long time ago. So I am going to focus on these older services today, but if you would like to read about newer ones. Let me know in the comments sections.

What is what? Life is life.

Before we go further, we have to know answer for questions like: “What is AI?” or “What is Machine Learning?”. You can find the answers in article How to distinguish types of AI .

But before we go to specific details there is a workflow how we work with ML models:

• Train - You have provide data and train the model
• Package - To be able to work with that
• Validate - You need to check if results are satisfying
• Deploy(eg. as web service) - When you have the model then you can make it publish for internal or external usage
• Monitor - When you you the model you can improve it

Machine Learning

There is one bit Azure Machine Learning service. It is a PaaS platform that include everything you need to build and use your own model. There are for example:

• Machine Learning Workspace - It it a top level ML resource in Azure
  • It is the main resource that manage the whole process
• Notebooks - It allow us to run Python or R script, but it contains also some example scripts
• Automated ML - A tool that can build ML model automatically
• Designer - A tool for creating ML models with user interface
• Datastore - To connect you other storages to ML tools
• Compute - to run model building

Summary

I know it was shorter material than usual, but I believe it will be expanded in the future. If you have any question, let me know in comments section below. If you want to be informed about new posts, subscribe.