Beginning Mlops with Mlflow: Deploy Models in Aws Sagemaker, Google Cloud, and Microsoft Azure

ISBN-13: 9781484265482 / Angielski / Miękka / 2020 / 330 str.

Sridhar Alla; Suman Kalyan Adari

Opis

Intermediate-Advanced user level

Spis treści

Chapter 1:   Getting Started: Data Analysis and Feature Engineering    

Chapter Goal: Establish the premise of the problem we want to solve with machine learning. Analyze several data sets and process them.  

No of pages - 30 pages 

Sub - Topics 

1.   Premise 

4.   Data analysis 

5.   Feature engineering 

 

Chapter 2:  Building a Machine Learning Model 

Chapter Goal: Build a machine learning model on a data set / several data sets that we processed the data for in chapter 4.

No of pages – 40 pages

Sub - Topics:  

1. Building the model 

2. Training and testing the model 

3.  Validation and optimizing 

 

Chapter 3:   What is MLOps? 

Chapter Goal: Introduce the reader to MLOps, various stages of automation in MLOps setups, automation with pipeline, and to CI/CD and CD Deployment.  

Pipelines for: source repo to deployment, prediction services, performance monitoring, etc 

Continuous Integration (source repo updated with new models), and Continuous Delivery (new models deployed). 

 No of pages – 40 pages  

Sub -Topics   

1. What is MLOps? 

2. MLOps setups 

3. Automation 

4. CI/CD – Continuous Integration & Delivery 

5. CD - Deployment 

 

Chapter 4:  Introduction to MlFlow

Chapter Goal: Introduce the reader to MLFlow and how to incorporate MLFlow into our ML training process (PyTorch, Keras, TensorFlow) 

 No of pages  – 30 pages 

Sub - Topics:  

1.  What is MLFlow?

2.  MLFlow in PyTorch

3.  MLFlow in Keras

4. MLFlow in TensorFlow

 

 Chapter 5:  Deploying in AWS – 40 pages 

Chapter Goal: Guide the reader through the process of deploying an MLOps setup on AWS SageMaker.    

-Description: The chapter will walk the reader through AWS SageMaker and help them deploy their MLOps setup (data processing scripts, model train, test, validation scripts) in AWS.

Chapter 6:  Deploying in Azure – 40 pages 

Chapter Goal:  Guide the reader through the process of deploying an MLOps setup on Microsoft Azure.

-Description: The chapter will walk the reader through Microsoft Azure and help them deploy their MLOps setup (data processing scripts, model train, test, validation scripts) in Azure.

 

Chapter 7:  Deploying in Google – 40 pages 

Chapter Goal:  Guide the reader through the process of deploying an MLOps setup on Google Cloud.

-Description: The chapter will walk the reader through Google Cloud and help them deploy their MLOps setup (data processing scripts, model train, test, validation scripts) in Google Cloud.

 

Appendix A:  a2ml – 20 pages 

Chapter Goal: This appendix chapter is optional and guides users through the process of deploying an MLOps setup using a2ml.  

-Description: The chapter will walk the reader through a2ml and help them deploy their MLOps setup (data processing scripts, model train, test, validation scripts) through a2ml.

O autorze

Sridhar Alla is the co-founder and CTO of Bluewhale, which helps big and small organizations build AI-driven big data solutions and analytics. He is a published author of books and an avid presenter at numerous Strata, Hadoop World, Spark Summit, and other conferences. He also has several patents filed with the US PTO on large-scale computing and distributed systems. He has extensive hands-on experience in several technologies, including Spark, Flink, Hadoop, AWS, Azure, Tensorflow, Cassandra, and others. He spoke on Anomaly Detection Using Deep Learning at Strata SFO in March of 2019 and at Strata London in October of 2019. He was born in Hyderabad, India and now lives in New Jersey, USA with his wife Rosie and daughter Evelyn. When he is not busy writing code, he loves to spend time with his family and also training, coaching, and organizing meetups. 

Suman Kalyan Adari is an undergraduate student pursuing a BS degree in computer science at the University of Florida. He has been conducting deep learning research in the field of cybersecurity since his freshman year, and has presented at the IEEE Dependable Systems and Networks workshop on Dependable and Secure Machine Learning held in Portland, Oregon, USA in June of 2019. He is passionate about deep learning, and specializes in its practical uses in various fields such as image recognition, anomaly detection, natural language processing, targeted adversarial attacks, and more. 

Blurb

Integrate MLOps principles into existing or future projects using MLFlow, operationalize your models, and deploy them in AWS SageMaker, Google Cloud, and Microsoft Azure. ​This book guides you through the process of data analysis, model construction, and training.

The authors begin by introducing you to basic data analysis on a credit card data set and teach you how to analyze the features and their relationships to the target variable. You will learn how to build logistic regression models in scikit-learn and PySpark, and you will go through the process of hyperparameter tuning with a validation data set. You will explore three different deployment setups of machine learning models with varying levels of automation to help you better understand MLOps. MLFlow is covered and you will explore how to integrate MLOps into your existing code, allowing you to easily track metrics, parameters, graphs, and models. You will be guided through the process of deploying and querying your models with AWS SageMaker, Google Cloud, and Microsoft Azure. And you will learn how to integrate your MLOps setups using Databricks.

You will:

• Perform basic data analysis and construct models in scikit-learn and PySpark
• Train, test, and validate your models (hyperparameter tuning)
• Know what MLOps is and what an ideal MLOps setup looks like
• Easily integrate MLFlow into your existing or future projects
• Deploy your models and perform predictions with them on the cloud