Update cookies preferences

E-book: Programming ML.NET

,
  • Format: 256 pages
  • Series: Developer Reference
  • Pub. Date: 03-Feb-2022
  • Publisher: Addison Wesley
  • Language: eng
  • ISBN-13: 9780137383627
Other books in subject:
  • Format - EPUB+DRM
  • Price: 38,60 €*
  • * the price is final i.e. no additional discount will apply
  • Add to basket
  • Add to Wishlist
  • This ebook is for personal use only. E-Books are non-refundable.
Programming ML.NETLarger Image
  • Format: 256 pages
  • Series: Developer Reference
  • Pub. Date: 03-Feb-2022
  • Publisher: Addison Wesley
  • Language: eng
  • ISBN-13: 9780137383627
Other books in subject:

DRM restrictions

  • Copying (copy/paste):

    not allowed

  • Printing:

    not allowed

  • Usage:

    Digital Rights Management (DRM)
    The publisher has supplied this book in encrypted form, which means that you need to install free software in order to unlock and read it.  To read this e-book you have to create Adobe ID More info here. Ebook can be read and downloaded up to 6 devices (single user with the same Adobe ID).

    Required software
    To read this ebook on a mobile device (phone or tablet) you'll need to install this free app: PocketBook Reader (iOS / Android)

    To download and read this eBook on a PC or Mac you need Adobe Digital Editions (This is a free app specially developed for eBooks. It's not the same as Adobe Reader, which you probably already have on your computer.)

    You can't read this ebook with Amazon Kindle

The expert guide to creating production machine learning solutions with ML.NET!

ML.NET brings the power of machine learning to all .NET developers— and Programming ML.NET helps you apply it in real production solutions. Modeled on Dino Esposito’s best-selling Programming ASP.NET, this book takes the same scenario-based approach Microsoft’s team used to build ML.NET itself. After a foundational overview of ML.NET’s libraries, the authors illuminate mini-frameworks (“ML Tasks”) for regression, classification, ranking, anomaly detection, and more. For each ML Task, they offer insights for overcoming common real-world challenges. Finally, going far beyond shallow learning, the authors thoroughly introduce ML.NET neural networking. They present a complete example application demonstrating advanced Microsoft Azure cognitive services and a handmade custom Keras network— showing how to leverage popular Python tools within .NET.

14-time Microsoft MVP Dino Esposito and son Francesco Esposito show how to:

  • Build smarter machine learning solutions that are closer to your user’s needs
  • See how ML.NET instantiates the classic ML pipeline, and simplifies common scenarios such as sentiment analysis, fraud detection, and price prediction
  • Implement data processing and training, and “productionize” machine learning–based software solutions
  • Move from basic prediction to more complex tasks, including categorization, anomaly detection, recommendations, and image classification
  • Perform both binary and multiclass classification
  • Use clustering and unsupervised learning to organize data into homogeneous groups
  • Spot outliers to detect suspicious behavior, fraud, failing equipment, or other issues
  • Make the most of ML.NET’s powerful, flexible forecasting capabilities
  • Implement the related functions of ranking, recommendation, and collaborative filtering
  • Quickly build image classification solutions with ML.NET transfer learning
  • Move to deep learning when standard algorithms and shallow learning aren’t enough
  • “Buy” neural networking via the Azure Cognitive Services API, or explore building your own with Keras and TensorFlow


With .NET 5's ML.NET and Programming ML.NET, any Microsoft .NET developer can solve serious machine learning problems, increasing their value and competitiveness in some of today's fastest-growing areas of software development. World-renowned Microsoft development expert Dino Esposito covers everything students need to know about ML.NET, the machine learning pipeline, and real-world machine learning solutions development.

  • Modeled on Esposito's popular Programming ASP.NET books
  • Use the same scenario-based approach Microsoft's team used to build the ML.NET framework itself
  • Discover ML.NET's dedicated mini-frameworks (“ML Tasks”) for specific classes of problems
  • Draw on Esposito's personal experience to apply these problems in the real world
  • Learn key concepts and realistic examples related to ML.NET neural networks
  • Leverage powerful Python-based machine learning tools in the .NET environment

Programming ML.NET will help students add machine learning and artificial intelligence to their tool belt, whether they have a background in these high-demand technologies or not.

Acknowledgments xv
Introduction xvii
Chapter 1 Artificially Intelligent Software
1(8)
How We Ended Up with Software
2(1)
The Formalization of Computing Machines
2(1)
The Engineering of Computing Machines
3(1)
The Birth of Artificial Intelligence
3(1)
Software as a Side Effect
4(1)
The Role of Software Today
4(1)
Automating Tasks
5(1)
Mirroring the Real World
5(1)
Empowering People
6(1)
Al Is Just Software
7(2)
Chapter 2 An Architectural Perspective of ML.NET
9(18)
Life Beyond Python
10(1)
Why Is Python So Popular in Machine Learning?
10(1)
Taxonomy of Python Machine Learning Libraries
11(2)
End-to-End Solutions on Top of Python Models
13(1)
Introducing ML.NET
13(1)
The Learning Pipeline in ML.NET
14(5)
Model Training Executive Summary
19(4)
Consuming a Trained Model
23(1)
Making the Model Callable from the Outside
23(1)
Other Deployment Scenarios
23(1)
From Data Science to Programming
24(1)
Summary
25(2)
Chapter 3 The Foundation of ML.NET
27(18)
On the Way to Data Engineering
27(1)
The Role of a Data Scientist
28(1)
The Role of a Data Engineer
29(1)
The Role of an ML Engineer
30(1)
The Data to Start From
30(1)
Making Sense of the Available Data
31(1)
Building a Data Processing Pipeline
32(3)
The Training Step
35(1)
Picking an Algorithm
36(1)
Measuring the Actual Value of an Algorithm
36(1)
Planning the Testing Phase
37(1)
A Look at the Metrics
38(1)
Consuming the Model from a Client Application
39(1)
Getting the Model File
39(1)
The Overall Project
39(1)
Making a Taxi Fare Prediction
40(2)
Scalability Concerns
42(1)
Devising an Adequate User Interface
42(1)
Summary
43(2)
Chapter 4 Prediction Tasks
45(28)
The Pipeline and the Chain of Estimators
46(1)
Data Views
46(1)
Transformers
47(1)
Estimators
47(1)
Pipelines
48(1)
The Regression ML Task
48(1)
General Aspects of ML Tasks
49(1)
Supported Regression Algorithms
49(3)
Supported Validation Techniques
52(2)
Using the Regression Task
54(1)
A Look at the Available Training Data
54(4)
Feature Engineering
58(2)
Accessing the Content of Datasets
60(2)
Composing the Training Pipeline
62(8)
The ML Devil's Advocate
70(1)
Simple and Linear Regression
70(1)
Nonlinear Regression
71(1)
Summary
71(2)
Chapter 5 Classification Tasks
73(26)
The Binary Classification ML Task
73(1)
Supported Algorithms
73(2)
Supported Validation Techniques
75(1)
Binary Classification for Sentiment Analysis
75(1)
A Look at the Available Training Data
75(4)
Feature Engineering
79(2)
Composing the Training Pipeline
81(4)
The Multiclass Classification ML Task
85(1)
Supported Algorithms
85(2)
Using the Multiclass Classification Task
87(1)
A Look at the Available Data
88(2)
Composing the Training Pipeline
90(6)
The ML Devil's Advocate
96(1)
The Many Faces of Classification
97(1)
Another Perspective on Sentiment Analysis
98(1)
Summary
98(1)
Chapter 6 Clustering Tasks
99(20)
The Clustering ML Task
99(1)
Unsupervised Learning
99(1)
A Look at the Available Training Data
100(4)
Feature Engineering
104(1)
Clustering Algorithms
105(4)
Composing the Training Pipeline
109(2)
Setting Up a Client Application
111(3)
The ML Devil's Advocate
114(1)
Clustering Is Always the First Step
114(1)
Unsupervised Reduction of the Dataset
115(2)
Summary
117(2)
Chapter 7 Anomaly Detection Tasks
119(22)
What Is an Anomaly?
119(1)
General Approaches to Detect Anomalies
120(1)
Time Series Data
120(2)
Statistical Techniques
122(1)
Machine Learning Approaches
123(2)
The Anomaly Detection ML Task
125(1)
A Look at the Available Training Data
125(3)
Composing the Training Pipeline
128(6)
Setting Up a Client Application
134(3)
The ML Devil's Advocate
137(1)
Predictive Maintenance
137(2)
Fraudulent Financial Operations
139(1)
Summary
140(1)
Chapter 8 Forecasting Tasks
141(18)
Predicting the Future
141(1)
Simple Forecasting Methods
141(1)
Mathematical Foundation of Forecasting
142(2)
Common Decomposition Algorithms
144(1)
The SSA Algorithm
144(2)
The Forecast ML Task
146(1)
A Look at the Available Data
146(2)
Composing the Training Pipeline
148(3)
Setting Up a Client Application
151(3)
The ML Devil's Advocate
154(1)
Not A Random Walk in the Park?
154(1)
Other Approaches to Time Series
155(1)
Energy Production Prediction
156(2)
Summary
158(1)
Chapter 9 Recommendation Tasks
159(16)
Inside Information Retrieval Systems
159(1)
The Basic Art of Ranking
160(1)
The Flexible Art of Recommendation
161(1)
The Delicate Art of Collaborative Filtering
162(1)
The ML Recommendation Task
163(1)
A Look at the Available Data
163(3)
Composing the Training Pipeline
166(4)
Setting Up a Client Application
170(2)
ML Devil's Advocate
172(1)
If You're Like Netflix
172(1)
What If You're Not Like Netflix?
173(1)
Summary
173(2)
Chapter 10 Image Classification Tasks
175(14)
Transfer Learning
175(1)
Popular Image Processing Neural Networks
176(1)
Other Image Neural Networks
176(1)
Transfer Learning via Composition
176(1)
The Transfer Learning Pattern in ML.NET
177(1)
Overall Purpose of the New Image Classifier
178(1)
A Look at the Available Data
178(2)
Composing the Training Pipeline
180(2)
Setting Up a Client Application
182(2)
The ML Image Classification Task
184(1)
The Image Classification API
184(1)
Using the Image Classification API
185(1)
The ML Devil's Advocate
186(1)
The Magic of the Human Brain
186(1)
Handcrafted Neural Networks
187(1)
Retraining
188(1)
Summary
188(1)
Chapter 11 Overview of Neural Networks
189(16)
Feed-forward Neural Networks
189(1)
Artificial Neurons
190(1)
Layers of the Network
191(2)
The Logistic Neuron
193(1)
Training a Neural Network
194(3)
More Sophisticated Neural Networks
197(1)
Stateful Neural Networks
197(2)
Convolutional Neural Networks
199(3)
Auto-Encoders
202(1)
Summary
203(2)
Chapter 12 A Neural Network to Recognize Passports
205(14)
Using Azure Cognitive Services
205(1)
Anatomy and Solution of the Problem
206(1)
Working with the ID Form Recognizer
207(3)
Crafting Your Own Neural Network
210(1)
Topology of the Neural Network
211(4)
Training Pains
215(1)
The ML Devil's Advocate
216(1)
Commodity Versus Vertical Solutions
216(1)
When Are Custom Solutions Inevitable?
217(1)
Summary
217(2)
Appendix A Model Explainability
219(5)
Software Intelligence
219(1)
The Super Theory of Artificial Intelligence
220(1)
Machine Learning Black Boxes
221(1)
Interpretability and Explainability
221(2)
Explainability Techniques
223(1)
Conclusion 224(1)
Index 225
Dino Esposito is CTO and co-founder of Crionet, a company that provides innovative software and technology to professional sports organizations. A 16-time Microsoft MVP, he has authored 20+ books, including Introducing Machine Learning; and the Microsoft Press best-seller Microsoft .NET: Architecting Applications for the Enterprise.

Francesco Esposito holds a degree in Mathematics, is the co-author of Introducing Machine Learning, and lives suspended between the depth of advanced mathematics and the intrigue of data science. He currently serves as the Head of Engineering and Data at Crionet. As an entrepreneur he founded Youbiquitous, a data analysis and software factory, and KBMS Data Force, a startup in Digital Therapy and intelligent healthcare.
More information about ebooks More information about ebooks
Permanent link: https://www.kriso.ee/db/97801373836276e.html
Keywords: