Hands-On Machine Learning with Azure

Book Description

Implement machine learning, cognitive services, and artificial intelligence solutions by leveraging Azure cloud technologies

Key Features

  • Learn advanced concepts in Azure ML and the Cortana Intelligence Suite architecture
  • Explore ML Server using SQL Server and HDInsight capabilities
  • Implement various tools in Azure to build and deploy machine learning models

Book Description

Implementing Machine learning (ML) and Artificial Intelligence (AI) in the cloud had not been possible earlier due to the lack of processing power and storage. However, Azure has created ML and AI services that are easy to implement in the cloud. Hands-On Machine Learning with Azure teaches you how to perform advanced ML projects in the cloud in a cost-effective way.

The book begins by covering the benefits of ML and AI in the cloud. You will then explore Microsoft's Team Data Science Process to establish a repeatable process for successful AI development and implementation. You will also gain an understanding of AI technologies available in Azure and the Cognitive Services APIs to integrate them into bot applications. This book lets you explore prebuilt templates with Azure Machine Learning Studio and build a model using canned algorithms that can be deployed as web services. The book then takes you through a preconfigured series of virtual machines in Azure targeted at AI development scenarios. You will get to grips with the ML Server and its capabilities in SQL and HDInsight. In the concluding chapters, you'll integrate patterns with other non-AI services in Azure.

By the end of this book, you will be fully equipped to implement smart cognitive actions in your models.

What you will learn

  • Discover the benefits of leveraging the cloud for ML and AI
  • Use Cognitive Services APIs to build intelligent bots
  • Build a model using canned algorithms from Microsoft and deploy it as a web service
  • Deploy virtual machines in AI development scenarios
  • Apply R, Python, SQL Server, and Spark in Azure
  • Build and deploy deep learning solutions with CNTK, MMLSpark, and TensorFlow
  • Implement model retraining in IoT, Streaming, and Blockchain solutions
  • Explore best practices for integrating ML and AI functions with ADLA and logic apps

Who this book is for

If you are a data scientist or developer familiar with Azure ML and cognitive services and want to create smart models and make sense of data in the cloud, this book is for you. You'll also find this book useful if you want to bring powerful machine learning services into your cloud applications. Some experience with data manipulation and processing, using languages like SQL, Python, and R, will aid in understanding the concepts covered in this book

Publisher Resources

Download Example Code

Table of Contents

  1. Title Page
  2. Copyright and Credits
    1. Hands-On Machine Learning with Azure
  3. Packt Upsell
    1. Why subscribe?
    2. Packt.com
  4. Contributors
    1. About the authors
    2. About the reviewer
    3. Packt is searching for authors like you
  5. Preface
    1. Who this book is for
    2. What this book covers
    3. To get the most out of this book
      1. Download the example code files
      2. Download the color images
      3. Conventions used
    4. Get in touch
      1. Reviews
  6. AI Cloud Foundations
    1. The importance of artificial intelligence
    2. The emergence of the cloud
      1. Essential cloud components for AI
    3. The Microsoft cloud – Azure
      1. Choosing AI tools on Azure
        1. Cognitive Services/bots
        2. Azure Machine Learning Studio
        3. ML Server
        4. Azure ML Services
        5. Azure Databricks
    4. Summary
  7. Data Science Process
    1. TDSP stages
      1. Business understanding
        1. Deliverable
      2. Data acquisition and understanding
        1. Deliverable
      3. Modeling
        1. Deliverable
      4. Deployment
        1. Deliverable
      5. Customer acceptance
        1. Deliverable
    2. Tools for TDSP
      1. IDEAR tool for R
      2. Automated modeling and reporting (AMAR) in R
    3. Summary
  8. Cognitive Services
    1. Cognitive Services for Vision APIs
    2. The Computer Vision API
    3. Face API
    4. Cognitive Services for Language APIs
      1. Text Analytics
    5. Cognitive Services for Speech APIs
      1. Speech to Text
    6. Cognitive Services for Knowledge APIs
      1. QnA Maker
    7. Cognitive Services for Search APIs
      1. Bing Visual Search
    8. Summary
    9. Reference
  9. Bot Framework
    1. What is a bot?
      1. Bot Builder SDK
      2. Bot Framework
      3. QnA Maker
      4. Bot Service
    2. Creating a bot with Bot Service
      1. LUIS application
    3. Summary
  10. Azure Machine Learning Studio
    1. Deploying an Azure AI Gallery template
    2. Building an experiment
      1. Importing and preprocessing data
      2. Choosing and configuring algorithms
      3. Feature selection
      4. Comparing models and parameters
    3. Deploying a model as a web service
      1. Creating a predictive experiment
      2. Deploying and testing a web service
    4. Summary
  11. Scalable Computing for Data Science
    1. Different scalable compute options in Azure
    2. Introduction to DSVMs
      1. Provisioning a DSVM
    3. DLVM
    4. Batch AI service
      1. Provisioning a Batch AI service
    5. ACI
    6. AKS
    7. Summary
  12. Machine Learning Server
    1. What is Microsoft ML Server?
      1. How to get started with Microsoft ML Server
        1. Configuring the DSVM
    2. Machine learning with Python
      1. Getting started with Python
        1. Set up your Python environment in Visual Studio
      2. Writing your own code with Python in Microsoft ML Server
      3. Walk-through: reading data in and out in Microsoft ML Server
      4. Introducing regression with Python in Microsoft ML Server
        1. More data visualization charts in Python and the Microsoft Machine Learning service
        2. Regression code walk-through with Python and Microsoft ML Server
      5. Analyzing results in machine learning models
        1. Measuring the fit of the model
        2. Cross validation
        3. Variance and bias
    3. Summary
  13. HDInsight
    1. R with HDInsight
    2. Getting started with Azure HDInsight and ML services
      1. Setup and configuration of HDInsight
        1. Basic configuration of HDInsight
        2. Storage options for Azure HDInsight
        3. Connect to the HDInsight cluster using SSH
        4. Accessing Microsoft ML Services on Azure HDInsight
    3. HDInsight and data analytics with R
      1. How do Azure Data Factory and HDInsight interact?
      2. Running queries on Azure HDInsight with ML Services
      3. RevoScaleR in Azure
      4. How can we read data into HDInsight using ML Services?
        1. What kind of analyzes can we do with R in ML Services on HDinsight?
      5. Reading data from files into Azure HDInsight ML Services
      6. Converting text and CSV files to the preferred XDF format
      7. Using the new XDF file in Microsoft ML Services
      8. XDF versus flat text files
      9. Reading data from SQL Server
        1. Connecting to a SQL Server database
        2. Extracting data from a table retrieving data from Microsoft SQL Server
      10. Installing R packages on Microsoft ML Services
      11. Analyzing and summarizing data in Microsoft ML Services
        1. Cross tabs and univariate statistics
        2. Working with cubes of data
        3. Grouping data using Microsoft ML Server and R
        4. Computing quantiles with R in Microsoft ML Server
        5. Logistic regression in Microsoft ML Services
        6. Predicting values with the model
      12. Visualizing data
        1. Creating histograms
        2. Creating line plots
    4. Enriching data for analysis
      1. rxDataSteps
    5. Summary
  14. Machine Learning with Spark
    1. Machine learning with Azure Databricks
      1. What challenges is Databricks trying to solve?
    2. Getting started with Apache Spark and Azure Databricks
      1. Creating a cluster
      2. Create a Databricks Notebook
    3. Using SQL in Azure Databricks
      1. Displaying data
    4. Machine Learning with HDInsight
      1. What is Spark?
    5. HDInsight and Spark
      1. The YARN operation system in Apache Spark
    6. Working with data in a Spark environment
      1. Using Jupyter Notebooks
    7. Configuring the data science virtual machine
      1. Running Spark MLib commands in Jupyter
        1. Data ingestion
        2. Data exploration
      2. Feature engineering in Spark
        1. Using Spark for prediction
      3. Loading a pipeline model and evaluating the test data
    8. Setting up an HDInsight cluster with Spark
      1. Provisioning an HDInsight cluster
    9. Summary
    10. Further references
  15. Building Deep Learning Solutions
    1. What is deep learning?
      1. Differences between traditional machine learning and deep learning
      2. Common Deep Learning Neural Networks (DNNs)
    2. Overview of the Azure Notebook service
      1. Pivot table formation with Azure Notebook
    3. Overview of Azure Deep Learning Virtual Machine toolkits
      1. Open source deep learning frameworks
        1. In-depth analysis of Microsoft deep learning tools 
        2. Overview of Microsoft CNTK
        3. The architecture building blocks of CNTK
          1. Concepts on CNTK
        4. Developing and deploying CNTK layers in the Azure Deep Learning VM to implement a neural network
          1. CNTK inputs and variables declaration
          2. CNTK variables section
          3. Data readers for CNTK
          4. Operations in CNTK
          5. Layers of the Microsoft CNTK 
          6. CNTK layer provision helpers
          7. CNTK modules for losses and error handling
          8. Input training models in CNTK
          9. Instantiating the Trainer object   
          10. Defining the training session object
          11. The CNTK testing model
        5. Deploying CNTK tools by using Azure Containers (Docker)
      2. Keras as a backend for Microsoft CNTK
    4. An overview of the Microsoft Machine Learning Library for Apache Spark (MMLSpark)
      1. Environment setup for MMLSpark
        1. Execution of MMLSpark notebooks using a Docker container 
        2. Azure HDInsight Spark cluster setup for MMLSpark
    5. Overview of TensorFlow on Azure 
      1. Simple computation graph on TensorFlow
        1. TensorFlow operations 
        2. Declaration of the TensorFlow placeholder 
        3. Neural Network Formation using TensorFlow 
      2. TensorFlow training 
        1. Execution of TensorFlow on Azure using Docker container services
        2. Running TensorFlow containers on an Azure Kubernetes Cluster (AKS)
      3. Other deep learning libraries  
    6. Summary
  16. Integration with Other Azure Services
    1. Logic Apps
      1. Triggers and actions
      2. Twitter sentiment analysis
      3. Adding language detection
    2. Azure Functions
      1. Triggers
      2. Blob-triggered function
    3. Azure Data Lake Analytics
      1. Developing with U-SQL
      2. U-SQL databases
      3. Simple format conversion for blobs
      4. Integration with Cognitive Services
    4. Azure Data Factory
      1. Datasets, pipelines, and linked services
      2. File format conversion
      3. Automate U-SQL scripts
      4. Running Databricks jobs
    5. Summary
  17. End-to-End Machine Learning
    1. Using the Azure Machine Learning SDK for E2E machine learning
    2. Summary
  18. Other Books You May Enjoy
    1. Leave a review - let other readers know what you think

Product Information

  • Title: Hands-On Machine Learning with Azure
  • Author(s): Thomas K Abraham, Parashar Shah, Jen Stirrup, Lauri Lehman, Anindita Basak
  • Release date: October 2018
  • Publisher(s): Packt Publishing
  • ISBN: 9781789131956