Wow! eBook
Practical Deep Learning at Scale with MLflow: Bridge the gap between offline experimentation and online production
Author(s): Yong Liu (Author)
- Publisher: Packt Publishing
- Publication Date: 8 July 2022
- Language: English
- Print length: 288 pages
- ISBN-10: 1803241330
- ISBN-13: 9781803241333
Book Description
Train, test, run, track, store, tune, deploy, and explain provenance-aware deep learning models and pipelines at scale with reproducibility using MLflow
Key Features
- Focus on deep learning models and MLflow to develop practical business AI solutions at scale
- Ship deep learning pipelines from experimentation to production with provenance tracking
- Learn to train, run, tune and deploy deep learning pipelines with explainability and reproducibility
Book Description
The book starts with an overview of the deep learning (DL) life cycle and the emerging Machine Learning Ops (MLOps) field, providing a clear picture of the four pillars of deep learning: data, model, code, and explainability and the role of MLflow in these areas.
From there onward, it guides you step by step in understanding the concept of MLflow experiments and usage patterns, using MLflow as a unified framework to track DL data, code and pipelines, models, parameters, and metrics at scale. You’ll also tackle running DL pipelines in a distributed execution environment with reproducibility and provenance tracking, and tuning DL models through hyperparameter optimization (HPO) with Ray Tune, Optuna, and HyperBand. As you progress, you’ll learn how to build a multi-step DL inference pipeline with preprocessing and postprocessing steps, deploy a DL inference pipeline for production using Ray Serve and AWS SageMaker, and finally create a DL explanation as a service (EaaS) using the popular Shapley Additive Explanations (SHAP) toolbox.
By the end of this book, you’ll have built the foundation and gained the hands-on experience you need to develop a DL pipeline solution from initial offline experimentation to final deployment and production, all within a reproducible and open source framework.
What you will learn
- Understand MLOps and deep learning life cycle development
- Track deep learning models, code, data, parameters, and metrics
- Build, deploy, and run deep learning model pipelines anywhere
- Run hyperparameter optimization at scale to tune deep learning models
- Build production-grade multi-step deep learning inference pipelines
- Implement scalable deep learning explainability as a service
- Deploy deep learning batch and streaming inference services
- Ship practical NLP solutions from experimentation to production
Who this book is for
This book is for machine learning practitioners including data scientists, data engineers, ML engineers, and scientists who want to build scalable full life cycle deep learning pipelines with reproducibility and provenance tracking using MLflow. A basic understanding of data science and machine learning is necessary to grasp the concepts presented in this book.
Table of Contents
- Deep Learning Life Cycle and MLOps Challenges
- Getting Started with MLflow for Deep Learning
- Tracking Models, Parameters, and Metrics
- Tracking Code and Data Versioning
- Running DL Pipelines in Different Environments
- Running Hyperparameter Tuning at Scale
- Multi-Step Deep Learning Inference Pipeline
- Deploying a DL Inference Pipeline at Scale
- Fundamentals of Deep Learning Explainability
- Implementing DL Explainability with MLflow
Editorial Reviews
Review
“Dr. Yong Liu tackles deep learning production issues head-on by showing you how to build robust and maintainable deep learning applications using MLflow. He brings a wealth of experience in production machine learning that shines through in every chapter of the book. He also presents all his material clearly. I hope that you enjoy learning from his experience and building groundbreaking applications using the latest techniques in deep learning.”
Matei Zaharia – Chief Technologist, Databricks, and Co-Creator of Mlflow
“Deep learning has quickly become a must have for industry grade AI systems. Dr. Liu’s book not only provides an in-depth tutorial on how to use DL, but most significantly, he covers the “development operations” of building, deploying and maintaining DL models in practical settings. While resources abound today on how to use a DL model, the knowledge of the entire lifecycle to build, test, deploy, log and manage DL models is rare. This book provides a deep look into the operations of DL model use with many code snippets, precise instructions and valuable examples.”
Steven Gustafson – CTO at Noonum, PhD in AI and Author
“Production machine learning systems are living, breathing systems. Because of the adaptability of the data, models, and code, they present unique challenge when it comes to deployment and maintenance. Dr. Yong Liu has written a very practical guide to implementing a production machine learning process using MLflow that any Python-based data science team can follow to handle all of the stages of the production machine learning deployment life cycle.”
Mike Tung, founder and CEO at Diffbot
“Deep learning models learn from screenshots of our moving reality. Scientists, researchers, and developers try to create the best possible models using those screenshots, accounting for biases to make the models as generalizable as possible. Most Deep Learning publications focus on model creation. Dr. Yong Liu’s book, “Practical Deep Learning at Scale with MLflow”,deals with the practicalities of the non-stationary nature of our world. Through the use of MLflow, Dr. Yong Liu helps you to answer very important questions, such as when and how a model in production should be changed to ensure better modeling of the current world without dips in performance―in the most efficient, timely, and cost-effective possible way―while keeping the model development process tractable and accountable by being able to explain the decision making involved.”
Manuel Reyes-Gomez, developer relations manager at NVIDIA
“Using deep learning at scale is becoming inevitable with ever-increasing volumes of data. This book is a must for anyone aiming to use deep learning at scale using MLFlow, from students to researchers and data scientists. The book introduces MLFlow and the challenges therein by organizing them into sections, and addressing them in detailed chapters, going from the fundamentals to deployment. This book exceeds expectations by also addressing some of the aspects of explainability in deep learning.”
Jeyan Thiyagalingam, Head of Scientific Machine Learning at Rutherford Appleton Laboratory, Science and Technology Facilities Council, UK
About the Author
Yong Liu has been working in big data science, machine learning, and optimization since his doctoral student years at the University of Illinois at Urbana-Champaign (UIUC) and later as a senior research scientist and principal investigator at the National Center for Supercomputing Applications (NCSA), where he led data science R&D projects funded by the National Science Foundation and Microsoft Research. He then joined Microsoft and AI/ML start-ups in the industry. He has shipped ML and DL models to production and has been a speaker at the Spark/Data+AI summit and NLP summit. He has recently published peer-reviewed papers on deep learning, linked data, and knowledge-infused learning at various ACM/IEEE conferences and journals.
