What is the primary purpose of AWS SageMaker?

The primary purpose of AWS SageMaker is to enable users to build, train, and deploy machine learning models. SageMaker provides a comprehensive set of tools and integrated environments that facilitate the entire machine learning lifecycle. This includes capabilities for data labeling, model training with built-in algorithms, hyperparameter tuning, and model deployment to production environments.

By simplifying and streamlining the process of developing machine learning applications, SageMaker allows data scientists and developers to focus on the modeling aspects rather than getting bogged down in the underlying infrastructure and complexities typically associated with machine learning projects. SageMaker also supports various frameworks and libraries, making it flexible for users with different preferences and use cases in machine learning.

The other options do not align with the core functionality of SageMaker. For instance, data backup and recovery is not part of SageMaker's primary offerings; instead, services like Amazon S3 and AWS Backup focus on that area. Real-time data streaming relates more closely to services such as Amazon Kinesis. Additionally, cost management and budgeting are functions typically handled by tools like AWS Budgets and AWS Cost Explorer, rather than a machine learning service like SageMaker.