Book Image

Mastering Azure Machine Learning

By : Christoph Körner, Kaijisse Waaijer
Book Image

Mastering Azure Machine Learning

By: Christoph Körner, Kaijisse Waaijer

Overview of this book

The increase being seen in data volume today requires distributed systems, powerful algorithms, and scalable cloud infrastructure to compute insights and train and deploy machine learning (ML) models. This book will help you improve your knowledge of building ML models using Azure and end-to-end ML pipelines on the cloud. The book starts with an overview of an end-to-end ML project and a guide on how to choose the right Azure service for different ML tasks. It then focuses on Azure Machine Learning and takes you through the process of data experimentation, data preparation, and feature engineering using Azure Machine Learning and Python. You'll learn advanced feature extraction techniques using natural language processing (NLP), classical ML techniques, and the secrets of both a great recommendation engine and a performant computer vision model using deep learning methods. You'll also explore how to train, optimize, and tune models using Azure Automated Machine Learning and HyperDrive, and perform distributed training on Azure. Then, you'll learn different deployment and monitoring techniques using Azure Kubernetes Services with Azure Machine Learning, along with the basics of MLOps—DevOps for ML to automate your ML process as CI/CD pipeline. By the end of this book, you'll have mastered Azure Machine Learning and be able to confidently design, build and operate scalable ML pipelines in Azure.

Related Content you might be interested in

Current Title:

Small book image
Mastering Azure Machine Learning
Christoph Körner | Kaijisse Waaijer
Apr, 2020 | 436 pages
Small book image
Azure Data Scientist Associate Certification Guide
Andreas Botsikas | Michael Hlobil
Dec, 2021 | 448 pages
Small book image
Azure Machine Learning Engineering
Megan Masanz | Sina Fakhraee | Balamurugan Balakreshnan
Jan, 2023 | 362 pages
Small book image
Automated Machine Learning with Microsoft Azure
Dennis Michael Sawyers
Apr, 2021 | 340 pages
Table of Contents (20 chapters)
Preface
Preface
About Mastering Azure Machine Learning
1
Section 1: Azure Machine Learning
Section 1: Azure Machine Learning
Free Chapter
2
1. Building an end-to-end machine learning pipeline in Azure
1. Building an end-to-end machine learning pipeline in Azure
Performing descriptive data exploration
Exploring common techniques for data preparation
Choosing the right ML model to train data
Optimization techniques
Deploying and operating models
Summary
3
2. Choosing a machine learning service in Azure
2. Choosing a machine learning service in Azure
Demystifying the different Azure services for ML
Azure Cognitive Services and Custom Vision
Azure Machine Learning with GUIs
Azure Machine Learning workspace
Summary
4
Section 2: Experimentation and Data Preparation
Section 2: Experimentation and Data Preparation
5
3. Data experimentation and visualization using Azure
3. Data experimentation and visualization using Azure
Preparing your Azure Machine Learning workspace
Visualizing high-dimensional data
Summary
6
4. ETL, data preparation, and feature extraction
4. ETL, data preparation, and feature extraction
Managing data and datasets in the cloud
Preprocessing and feature engineering with Azure Machine Learning DataPrep
Summary
7
5. Azure Machine Learning pipelines
5. Azure Machine Learning pipelines
Benefits of pipelines for ML workflows
Building and publishing an ML pipeline
Integrating pipelines with other Azure services
Summary
8
6. Advanced feature extraction with NLP
6. Advanced feature extraction with NLP
Understanding categorical data
Building a simple bag-of-words model
Leveraging term importance and semantics
Implementing end-to-end language models
Summary
9
Section 3: Training Machine Learning Models
Section 3: Training Machine Learning Models
10
7. Building ML models using Azure Machine Learning
7. Building ML models using Azure Machine Learning
Working with tree-based ensemble classifiers
Training an ensemble classifier model using LightGBM
Summary
11
8. Training deep neural networks on Azure
8. Training deep neural networks on Azure
Introduction to deep learning
Training a CNN for image classification
Summary
12
9. Hyperparameter tuning and Automated Machine Learning
9. Hyperparameter tuning and Automated Machine Learning
Hyperparameter tuning to find the optimal parameters
Finding the optimal model with Azure Automated Machine Learning
Summary
13
10. Distributed machine learning on Azure
10. Distributed machine learning on Azure
Exploring methods for distributed ML
Using distributed ML in Azure
Summary
14
11. Building a recommendation engine in Azure
11. Building a recommendation engine in Azure
Introduction to recommender engines
Content-based recommendations
Collaborative filtering—a rating-based recommendation engine
Combining content and ratings in hybrid recommendation engines
Automatic optimization through reinforcement learning
Summary
15
Section 4: Optimization and Deployment of Machine Learning Models
Section 4: Optimization and Deployment of Machine Learning Models
16
12. Deploying and operating machine learning models
12. Deploying and operating machine learning models
Deploying ML models in Azure
Building a real-time scoring service
Implementing a batch scoring pipeline
Inference optimizations and alternative deployment targets
Monitoring Azure Machine Learning deployments
Summary
17
13. MLOps—DevOps for machine learning
13. MLOps—DevOps for machine learning
Ensuring reproducible builds and deployments
Validating your code, data, and models
Summary
18
14. What's next?
14. What's next?
Understanding the importance of data
The future of ML is automated
Change is the only constant – preparing for change
Focusing first on infrastructure and monitoring
Controlled rollouts and A/B testing
Summary
19
Index
Index

Summary

In this chapter, you learned how to build a classical ML model in Azure Machine Learning.

You learned about decision trees, a popular technique for various classification and regression problems. The main strengths of decision trees are that they require little data preparation as they work well on categorical data and different data distributions. Another important benefit is their interpretability, which is especially important for business decisions and users. This helps you to understand when a decision-tree-based ensemble predictor is appropriate to use.

However, we also learned about a set of weaknesses, especially regarding overfitting and poor generalization. Luckily, tree-based ensemble techniques such as bagging (bootstrap aggregation) and boosting help to overcome these problems. While bagging has popular methods such as random forests that parallelize very well, boosting—especially gradient boosting—has efficient implementations such as XGBoost...

Previous Section
End of Chapter 10
Next Chapter