Hands-On Machine Learning with ML.NET

Create, train, and evaluate various machine learning models such as regression, classification, and clustering using ML.NET, Entity Framework, and ASP.NET Core

▶What You Will Learn
⦁ Understand the framework, components, and APIs of ML.NET using C#
⦁ Develop regression models using ML.NET for employee attrition and file classification
⦁ Evaluate classification models for sentiment prediction of restaurant reviews
⦁ Work with clustering models for file type classifications
⦁ Use anomaly detection to find anomalies in both network traffic and login history
⦁ Work with ASP.NET Core Blazor to create an ML.NET enabled web application
⦁ Integrate pre-trained TensorFlow and ONNX models in a WPF ML.NET application for image classification and object detection

▶Key Features
⦁ Get well-versed with the ML.NET framework and its components and APIs using practical examples
⦁ Learn how to build, train, and evaluate popular machine learning algorithms with ML.NET offerings
⦁ Extend your existing machine learning models by integrating with TensorFlow and other libraries

▶Who This Book Is For
If you are a .NET developer who wants to implement machine learning models using ML.NET, then this book is for you. This book will also be beneficial for data scientists and machine learning developers who are looking for effective tools to implement various machine learning algorithms. A basic understanding of C# or .NET is mandatory to grasp the concepts covered in this book effectively.

▶What this book covers
⦁ Chapter 1, Getting Started with Machine Learning and ML.NET, talks about what machine learning is and how important machine learning is in our society today. It also introduces ML.NET and talks in more detail about getting started with it after learning about the concepts of machine learning and how they relate.

⦁ Chapter 2, Setting Up the ML.NET Environment, talks in more detail about getting started with ML.NET, continuing the overview of machine learning and how ML.NET can assist in both developing and running models in both new and existing applications. You will ensure your development environment is set up and the chapter ends with a simple pretrained model in a console application to demonstrate that you are ready to proceed with the training.

⦁ Chapter 3, Regression Model, talks about using a regression and logistic regression model in ML.NET in addition to the math and what problems these models can help to solve. In addition, the chapter provides a step-by-step explanation of how to create and work with both a regression model and a logistic regression model in ML.NET. The end of the chapter details a quick console application using the dataset and both the models in ML.NET.

⦁ Chapter 4, Classification Model, talks about using the classifications trainer models in ML.NET and what problems a classification model can help to solve. For this chapter, we will create two applications to demonstrate the classification trainer support in ML.NET. The first predicts whether a car is of good value based on the several attributes and comparative prices using the FastTree trainer that ML.NET provides. The second application takes email data (Subject, Body, Sender) with the SDCA trainer in ML.NET to classify the email as an Order, Spam or Friend. Through these applications, you will also learn how to evaluate classification models.

⦁ Chapter 5, Clustering Model, talks about using the k-means clustering trainer in ML.NET in addition to what problems a clustering model can help to solve. In this chapter, we will use the k-means cluster trainer that ML.NET provides in order to create an example application that will classify files as either executables, documents, or scripts. In addition, you will learn how to evaluate clustering models in ML.NET.

⦁ Chapter 6, Anomaly Detection Model, talks about using an anomaly detection model in ML.NET in addition to what problems an anomaly detection model can help to solve. For this chapter, we will create two example applications. The first uses ML.NET with SSA to detect Network Traffic anomalies, while the second example uses ML.NET with PCA to detect anomalies in a series of user logins. With these applications, we will also look at how you can evaluate your anomaly detection model once trained.

⦁ Chapter 7, Matrix Factorization Model, talks about using a matrix factorization model in ML.NET in addition to the math and what problems a matrix factorization model can help to solve. In this chapter, we will create a music recommendation application using the matrix factorization trainer that ML.NET provides. Using several data points this recommendation engine will recommend music based on the training data provided to the model. In addition, after creating this application we will learn how to evaluate a matrix factorization model in ML.NET.

⦁ Chapter 8, Using ML.NET with .NET Core and Forecasting, covers a real-world application utilizing .NET Core and utilizes both a regression and time series model to demonstrate forecasting on stock shares.

⦁ Chapter 9, Using ML.NET with ASP.NET Core, covers a real-world application utilizing ASP.NET with a frontend to upload a file to determine whether it is malicious or not. This chapter focuses on using a binary classifier and how to integrate it into an ASP.NET application.

⦁ Chapter 10, Using ML.NET with UWP, covers a real-world application utilizing UWP and ML.NET. The application will utilize ML.NET to classify whether the web page content is malicious. The chapter will also cover UWP application design and MVVM briefly to give a true production-ready sample app to build on or adapt to other applications for using UWP with ML.NET.

⦁ Chapter 11, Training and Building Production Models, covers training a model at scale with all of the considerations, along with the proper training of a production model using the DMTP project. The lessons learned include obtaining proper training sets (diversity being key), proper features, and the true evaluation of your model. The focus of this chapter is on tips, tricks, and best practices for training production-ready models.

⦁ Chapter 12, Using TensorFlow with ML.NET, talks about using a pre-trained TensorFlow model with ML.NET to determine whether a car is in a picture or not with a UWP application.

⦁ Chapter 13, Using ONNX with ML.NET, talks about using a pre-trained ONNX model with ML.NET in addition to the value added by taking a pre-existing ONNX format model into ML.NET directly.