Integrate an Azure Machine Learning endpoint in your PowerApp

In this tutorial, you will use PowerApps and PowerAutomate to quickly create an app that is using a model deployed on ACI with Azure Machine Learning. There are two parts to this tutorial:

• Part 1: Creating a PowerAutomate flow to call upon the model after we click on a button in our PowerApp
• Part 2: Creating the PowerApp to fill in the data that will be sent to the model after clicking on a button and receiving the response/prediction from the model

Contents

This tutorial only covers how to integrate your already deployed model in PowerApps and PowerAutomate. The tutorial is based on a model that has been created and deployed in the DP-100: Designing and Implementing Data Science Solutions on Microsoft Azure course. Find the DP-100 instructions here.

If you haven't participated in the full DP-100 course, these are the labs you need to complete before you can do this tutorial:

• Lab 1A
• Lab 1B: Task 1 only (installing SDK and cloning Git Repo)
• Lab 7A

Resources

This tutorial makes use of the following technologies:

• Azure Machine Learning:
  • Documentation
  • Learning Path on Microsoft Learn
• A deployed model on Azure Container Instances:
  • Documentation
• Power Automate:
  • Documentation
  • Short introduction on Microsoft Learn
  • Learn how to create your own flows on Microsoft Learn
• PowerApps:
  • Documentation
  • Short introduction on Microsoft Learn
  • Learn how to create your own apps on Microsoft Learn