Preface

Implementing a product based on machine learning can be a laborious task. There is a general need to reduce the friction between different steps of the machine learning development life cycle and between the teams of data scientists and engineers that are involved in the process.

Machine learning practitioners such as data scientists and machine learning engineers operate with different systems, standards, and tools. While data scientists spend most of their time developing models in tools such as Jupyter Notebook, when running in production, the model is deployed in the context of a software application with an environment that's more demanding in terms of scale and reliability.

In this book, you will be introduced to MLflow and machine learning engineering practices that will aid your machine learning life cycle, exploring data acquisition, preparation, training, and deployment. The book's content is based on an open interface design and will work with any language or platform. You will also gain benefits when it comes to scalability and reproducibility.

By the end of this book, you will be able to comfortably deal with setting up a development environment for models using MLflow, framing your machine learning problem, and using a standardized framework to set up your own machine learning systems. This book is also particularly handy if you are implementing your first machine learning project in production.

Who this book is for

This book is geared toward software, machine learning, and data science professionals or enthusiasts who want to explore the engineering side of machine learning systems in production. Machine learning practitioners will be able to put their knowledge to work with this practical guide to MLflow. The book takes a hands-on approach to implementation and associated methodologies that will have you up and running with MLflow in no time. The basic requirements for this book are experience in Python programming and knowledge of the Bash terminal and commands.

What this book covers

Chapter 1, Introducing MLflow, will be an overview of the different features of MLflow, guiding you in installing and exploring the core features of the platform. After reading this chapter, you will be able to install and operate your MLflow environment locally.

Chapter 2, Your Machine Learning Project, introduces the focus of the book. The approach of this book is to work through a practical business case, namely stock market prediction, and, through this use case, explore all the different features of MLflow. A problem-framing framework will be used to get you deeply familiar with the example used in the book. A sample pipeline will be created for use in the remainder of the book.

Chapter 3, Your Data Science Workbench, helps you understand how to use MLflow to create your local environment so that you can develop your machine learning projects locally using all the different features provided by MLflow.

Chapter 4, Experiment Management in MLflow, is where you will gain practical experience of stock prediction by creating different models and comparing the metrics of different runs in MLflow. You will be guided as to how to deploy a tracking server so that many machine learning practitioners can share metrics and improve a model.

Chapter 5, Managing Models with MLflow, looks at the different features for model creation in MLflow. Built-in models, such as PyTorch and TensorFlow models, will be covered alongside custom models not available in MLflow. A model life cycle will be introduced alongside the Model Registry feature of MLflow.

Chapter 6, Introducing ML Systems Architecture, talks about the need to architect machine learning systems properly and how MLflow fits in the picture of an end-to-end machine learning system.

Chapter 7, Data and Feature Management, introduces data and feature management. The importance of feature generation will be made clear, as will how to use feature streams to log model results with MLflow.

Chapter 8, Training Models with MLflow, is where the complete training pipeline infrastructure will be described and developed for the problem at hand, with the use of MLflow-specific features.

Chapter 9, Deployment and Inference with MLflow, is where an end-to-end deployment infrastructure for our machine learning system, including the inference component, will be exposed using the API and batch features of MLflow. The cloud-enabled features of MLflow will also be described.

Chapter 10, Scaling Up Your Machine Learning Workflow, covers integration with high-performance/big data libraries that allow MLflow systems to scale for large volumes of data.

Chapter 11, Performance Monitoring, explores the important area of machine learning operations and how to ensure a smooth ride for the production systems developed in the book using best practices and operational patterns.

Chapter 12, Advanced Topics with MLFlow, presents advanced case studies with complete MLflow pipelines. The case studies use different types of models from the ones looked at in the rest of the book to ensure a breadth of feature coverage for MLflow.

To get the most out of this book

Ideally, before getting started with the book, you should have a good grasp of the Python programming language and should have already created basic machine learning models. One introductory course in machine learning will help contextualize the concepts discussed in the book.

If you are using the digital version of this book, we advise you to type the code yourself or access the code from the book's GitHub repository (a link is available in the next section). Doing so will help you avoid any potential errors related to the copying and pasting of code.

Download the example code files

You can download the example code files for this book from GitHub at https://github.com/PacktPublishing/Machine-Learning-Engineering-with-MLflow. If there's an update to the code, it will be updated in the GitHub repository.

We also have other code bundles from our rich catalog of books and videos available at https://github.com/PacktPublishing/. Check them out!

Download the color images

We also provide a PDF file that has color images of the screenshots and diagrams used in this book. You can download it here: https://static.packt-cdn.com/downloads/9781800560796_ColorImages.pdf

Conventions used

There are a number of text conventions used throughout this book.

Code in text: Indicates code words in text, database table names, folder names, filenames, file extensions, pathnames, dummy URLs, user input, and Twitter handles. Here is an example: "The model.pkl file contains a serialized version of the model"

A block of code is set as follows:

import mlflow
from sklearn.linear_model import LogisticRegression
mlflow.sklearn.autolog()
with mlflow.start_run():
    clf = LogisticRegression()
    clf.fit(X_train, y_train)

Any command-line input or output is written as follows:

docker build -t stockpred -f dockerfile

Bold: Indicates a new term, an important word, or words that you see onscreen. For instance, words in menus or dialog boxes appear in bold. Here is an example: "MLflow is an open source platform for the machine learning (ML) life cycle"

Tips or important notes

Appear like this.

Get in touch

Feedback from our readers is always welcome.

General feedback: If you have questions about any aspect of this book, email us at [e-mail] and mention the book title in the subject of your message.

Errata: Although we have taken every care to ensure the accuracy of our content, mistakes do happen. If you have found a mistake in this book, we would be grateful if you would report this to us. Please visit www.packtpub.com/support/errata and fill in the form.

Piracy: If you come across any illegal copies of our works in any form on the internet, we would be grateful if you would provide us with the location address or website name. Please contact us at [e-mail] with a link to the material.

If you are interested in becoming an author: If there is a topic that you have expertise in and you are interested in either writing or contributing to a book, please visit authors.packtpub.com.

Share Your Thoughts

Once you've read Machine Learning Engineering with MLflow, we'd love to hear your thoughts! Scan the QR code below to go straight to the Amazon review page for this book and share your feedback.

https://packt.link/r/1-800-56079-6

Your review is important to us and the tech community and will help us make sure we're delivering excellent quality content.