The training dataset includes both categorical and numerical features. To maximize model accuracy with the least operational overhead, how should the ML engineer prepare these features?

Amazon SageMaker Data Wrangler is a purpose-built tool within SageMaker for data preparation and feature engineering. It offers a visual interface and a wide array of transformations, including one-hot encoding, label encoding, and ordinal encoding, to convert categorical features into numerical representations suitable for machine learning models. This approach maximizes model accuracy by providing well-prepared data and minimizes operational overhead due to Data Wrangler's integrated and managed nature. Converting categorical features to numeric is essential because most machine learning algorithms require numerical input. Converting numeric features to categorical bins (discretization) is a valid technique but not the primary or most impactful step for maximizing accuracy when dealing with mixed data types. AWS Glue is a powerful ETL service but is more general-purpose; Data Wrangler is specialized for ML data preparation, offering a more streamlined experience for this specific task.