Understanding AutoML

Azure ML and AutoML may both be new concepts to you. If you do most of your work in Power BI, you may only use these tools occasionally. Multiple books can be dedicated to either of these concepts, which is why we'll cover the bare necessities for data analysts here.

So, why do we want to learn about AutoML? Throughout this book, we have explored many features and services that offer pretrained models that are ready to use. There is no need to train them, nor to have the data-science expertise to create models from scratch.

Pretrained models are ideal for common scenarios that many organizations face; for example, one model trained to recognize faces can be used for many different applications. However, if you want to have a forecasting model to predict the demand of your products based on your advertisement strategies to plan the supply, a generic model may not be the right fit for you.

It's when you need the model to be trained and tuned...