Machine Learning Infrastructure and Best Practices for Software Engineers

Take your machine learning software from a prototype to a fully fledged software system

Created by Miroslaw Staron

Explore how to transform your machine learning prototypes into robust, scalable software systems. Learn practical strategies for building, testing, and deploying ML pipelines that meet real-world demands. Gain insights into best practices for data management, infrastructure, and ethical considerations.

Packt | Jan 2024 | 346 min

Start Trial

LevelExpert

CategoriesSoftware Engineering and Architecture, Deep Learning Architectures and Frameworks

1Machine Learning Compared to Traditional Software
2Elements of a Machine Learning Software System
3Data in Software Systems – Text, Images, Code, Features
4Data Acquisition, Data Quality and Noise
5Quantifying and Improving Data Properties
6Types of Data in ML Systems
7Feature Engineering for Numerical and Image Data
8Feature Engineering for Natural Language Data
9Types of Machine Learning Systems – Feature-Based and Raw Data Based (Deep Learning)
10Training and evaluation of classical ML systems and neural networks
11Training and evaluation of advanced algorithms – deep learning, autoencoders, GPT-3
12Designing machine learning pipelines (MLOps) and their testing
13Designing and implementation of large scale, robust ML software – a comprehensive example
14Ethics in data acquisition and management
15Ethics in machine learning systems
16Integration of ML systems in ecosystems
17Summary and where to go next

What You Will Learn

You will gain hands-on experience by working through real-world scenarios that illustrate each concept. Practical examples and proven techniques guide you from initial design choices to deployment and maintenance. Along the way, you will learn to apply both foundational and cutting-edge practices used by experienced professionals.

Key Features

Build scalable machine learning pipelines ready for production environments
Implement best practices for data quality, testing, and validation
Address ethical risks and ensure responsible deployment of ML systems

Target Audience

Ideal for software engineers and machine learning practitioners with some experience who want to turn prototypes into reliable products. If you are looking to improve your understanding of scaling, infrastructure, and ethical challenges in ML systems, this course will help you build the skills needed to deliver robust solutions.

Related courses