Pragmatic Programmer for Machine Learning

Gives a holistic approach to machine learning and data science applications, from design to deployment and quality assurance, as an overarching cyclical process; Bridges machine learning and software engineering to build a shared set of best practices useful to both academia and the industry; Discusses deployment options for different types of models and data to help practitioners reason and make informed choices. Emphasizes the role of coding standards and software architecture alongside statistical rigor to implement reproducible and scalable machine learning modelsKey Features:

A complete guide to software engineering for machine learning and data science applications, from choosing the right hardware to analysing algorithms and designing scalable architectures. Surveys the state of the art of the software and frameworks used to build and run machine learning applications, comparing and contrasting their trade-offs. Comes with a complete case study in natural language understanding which illustrates the principles and the tools covered in the book. Code available from GitHub. Provides a multi-disciplinary view of how traditional software learning practices can be integrated with the workflows of domain experts and the unique characteristics of software in which data play a central role.