What does Amazon SageMaker Autopilot automate?

Amazon SageMaker Autopilot automates the process of creating, training, and tuning machine learning models. This service allows users to simply provide their data, after which Autopilot handles the various stages involved in model development. It analyzes the datasets, performs feature engineering, and selects the appropriate algorithms for training.

By automating these processes, SageMaker Autopilot enables users—regardless of their level of expertise in machine learning—to effectively build and optimize predictive models quickly and efficiently. This is particularly beneficial in case the users do not have a deep understanding of machine learning techniques or the nuances involved in model selection and hyperparameter tuning.

In contrast, while data preprocessing and cleaning, deployment of machine learning applications, and real-time monitoring of model performance are vital components of the machine learning lifecycle, they are not the primary functions of SageMaker Autopilot. The service focuses on automating the model creation and training aspect, leaving the other tasks to potentially require different services or tools within the AWS ecosystem.