Design Patterns für Machine Learning

Proven practical solutions for complex machine learning tasks. Covers all phases of the ML product pipeline. Clearly structured layout that ensures concepts and relationships are quickly understood. Focus on TensorFlow, but also transferable to PyTorch projects.

The design patterns in this book showcase tried-and-true methods and solutions for recurring tasks in machine learning. The authors, three machine learning experts at Google, describe established approaches to assist data scientists and data engineers in solving common problems throughout the ML process. The patterns consolidate the experiences of hundreds of experts and provide simple, accessible best practices.

In this book, you will find detailed explanations of 30 patterns covering the following topics: data and problem representation, operationalization, repeatability, reproducibility, flexibility, explainability, and fairness. Each pattern includes a description of the problem, a variety of possible solutions, and recommendations for selecting the best technique for your situation.