Feature Engineering

This introductory module provides an overview of the course and its objectives.

This module introduces Vertex AI Feature Store, covering its benefits, terminology, concepts, data model, creation process, and serving features in batch and online modes.

This module explores the process of converting raw data into useful features for machine learning models. Topics include identifying good features, ensuring feature availability at prediction time, and proper feature representation.

This module covers feature engineering in BigQuery ML and Keras, including basic and advanced techniques like feature crosses, bucketization, and temporal features. Includes hands-on labs for practical application.

This module explores Dataflow and Apache Beam for building and running preprocessing and feature engineering pipelines.

This module focuses on feature crosses in modern machine learning, including their recognition and application. Students learn about discretization and practical implementation using TensorFlow Playground.

This module introduces TensorFlow Transform (tf.Transform) for preprocessing data, covering analyze and transform phases, and supporting serving capabilities.

This concluding module provides a comprehensive summary of the Feature Engineering course, including review materials and resources.