Challenge 6 - Building Models in Azure Machine Learning Designer

Prerequisites

  1. Challenge 5 - Building Machine Learning in Power BI AutoML should be done successfully.

Introduction

Adventure Works marketing team loves the insights they’ve been getting from the Bike Buyer predictive model. They love it so much, they’d like to apply it more broadly and use it from other systems. Since the current model can only be used by Power BI, you’ve been asked to help Adventure Works recreate the model for use outside of Power BI. For our first step into this process lets use the Azure Machine Learning Studio Designer. In subsequent labs we will improve on this even more and learn new tools to rapidly implement ML Models.

Success Criteria

  1. A published BikeBuyer model endpoint hosted in Azure ML
  2. Demonstrate usage of the API
  3. Create a new dataflow in Power BI to leverage the published API

Hints

  1. Have a closer look at the columns in BikeBuyerTraining and Prospective Buyer. What is different about these two tables? Are there any differences that could be preventing you from using your model?
  2. If you don’t want to write code to call the REST API, Postman is a great tool. (http://getpostman.com)

Learning resources

   
Description Links
Azure Machine Learning Main Docs https://docs.microsoft.com/en-us/azure/machine-learning/
Tutorial: Predict automobile price with the designer (preview) https://docs.microsoft.com/en-us/azure/machine-learning/tutorial-designer-automobile-price-train-score

[Next challenge = None!!! You Are DONE!!! Congrats!!!]