Time Series Analysis on AWS

Chapter 4: Training a Predictor with AutoML

In Chapter 3, Creating a Project and Ingesting Your Data, you learned how a raw dataset obtained from a public data source could be prepared and ingested in a suitable format for Amazon Forecast. Additionally, you created your first forecasting project (which, in the service, is called a dataset group) and populated it with three datasets. These datasets included the very time series that you want to forecast and several related time series that, intuitively, will lend some predictive power to the model you are going to train.

In this chapter, you will use these ingested datasets to train a forecasting model in Amazon Forecast; this construct is called a Predictor. Following this, you will learn how ...