Summary

In this chapter, we explored Azure ML and how it can help us train and deploy models so that we can integrate them with Power BI. Throughout this chapter, we used the Azure ML Designer and Azure ML Studio to explore the process through the UI. All these steps can be replicated using the Python SDK or the Azure CLI extension for Azure ML if you want to use your training scripts and prefer to work with code. To get predictions in Power BI, you can choose to generate batch predictions upstream or deploy your model to a real-time endpoint to use the Azure Machine Learning feature in Power Query Editor. In the next chapter, we'll go over some things you should consider when you're training and deploying models so that you can use them responsibly.