Machine Learning Infrastructure and Best Practices for Software Engineers: Take your machine learning software from a prototype to a fully fledged software system,1st Edition

The book “Machine Learning Infrastructure and Best Practices for Software Engineers” gives software engineers all the necessary details on converting the machine learning pipelines proto-typed into fully-fledged software products. This is done to familiarize the reader with the difference between conventional software and software developed with ML and, therefore, the latter's peculiarities. It also covers initial best practices, such as choosing, growing, and evaluating algorithms.

As the book goes through, it dives into the required structure elements for large-scale machine learning systems. The readers will be informed on how high-quality data sources can be identified and acquired to foster reliability and the performance of the obtained software. The book ‘’Machine Learning Infrastructure and Best Practices for Software Engineers’’ also covers the issues related to maintaining the quality of the machine learning pipeline at runtime In addition, the methods that help to keep the software developed as a prototype to be robust and scalable after being implemented in the production mode are also explained.

That is why the book's final chapter is devoted to the ethical issues emerging when implementing big-scale machine learning systems. It gives recommendations on how ethical risk must be evaluated and managed to allow engineers to develop ethical software.

Machine Learning Infrastructure and Best Practices for Software Engineers Table of Contents:

1. Machine Learning Compared to Traditional Software
2. Elements of a Machine Learning Software System
3. Data in Software Systems - Text, Images, Code, Features
4. Data Acquisition, Data Quality and Noise
5. Quantifying and Improving Data Properties
6. Types of Data in ML Systems
7. Feature Engineering for Numerical and Image Data
8. Feature Engineering for Natural Language Data
9. Types of Machine Learning Systems - Feature-Based and Raw Data-Based (Deep Learning)
10. Training and evaluation of classical ML systems and neural networks
11. Training and evaluation of advanced algorithms - deep learning, autoencoders, GPT-3
12. Designing machine learning pipelines (MLOps) and their testing
13. Designing and implementation of large-scale, robust ML software - a comprehensive example
14. Ethics in data acquisition and management

Who is this course for?

1. Machine Learning Engineers: This book will assist you in devising better software and comprehending the problems of scaling up and why it is important to know them.
2. Software Engineers: Benefit from best practices to ensure that your products will be highly robust, highly reliable, and highly innovative.
3. Decision Makers: Get relevant knowledge concerning what the well-designed machine learning software product should possess.

Click on the links below to Download Machine Learning Infrastructure and Best Practices for Software Engineers!