What this book covers
Chapter 1, Data Science, Notebooks, and Kernels, introduces the idea of data science and interactive notebooks – and the scenarios notebooks help developers, data analysts, and data scientists tackle.
Chapter 2, Exploring Polyglot Notebooks, introduces the Polyglot Notebooks environment and .NET Interactive kernel to the reader by showing how interactive code cells and variable sharing work.
Chapter 3, Getting Data and Code into Your Notebooks, covers different ways of pulling data into a notebook environment including importing from CSV and TSV files, executing SQL and KQL queries, and pulling data in from external APIs.
Chapter 4, Working with Tabular Data and DataFrames, explores data analysis, data cleaning, and feature engineering using the ML.NET DataFrame object.
Chapter 5, Visualizing Data, shows how variable distributions can be explored, both in terms of descriptive statistics and in terms of various distribution plots in Plotly.NET and ScottPlot.
Chapter 6, Variable Correlations, takes data visualization to the next level by exploring how variable relationships can be visualized and introduces the idea of correlation scores and correlation matrixes.
Chapter 7, Classification Experiments with ML.NET AutoML, introduces machine learning through ML.NET by using ML.NET’s automated machine learning capabilities to perform our first classification experiment.
Chapter 8, Regression Experiments with ML.NET AutoML, expands our exploration into ML.NET by covering the prediction of numerical values with regression.
Chapter 9, Beyond AutoML: Pipelines, Trainers, and Transforms, moves away from automated machine learning and shows how ML.NET can be customized by using pipelines, transforms, model trainers, and hyperparameter tuning.
Chapter 10, Deploying Machine Learning Models, concludes our exploration of ML.NET by showing how models can be saved and loaded and embedded into other applications, such as an ASP .NET web application.
Chapter 11, Generative AI in Polyglot Notebooks, introduces generative AI and prompt engineering by using Azure OpenAI models for chat completions, image generation, and text embeddings.
Chapter 12, AI Orchestration with Semantic Kernel, expands on our generative AI knowledge by introducing retrieval-augmented generation (RAG), AI orchestration, and Microsoft Semantic Kernel to achieve complex tasks with AI agents.
Chapter 13, Enriching Documentation with Mermaid Diagrams, shows how Mermaid diagrams serve as simple maintainable markdown-based diagrams that help illustrate common engineering workflows.
Chapter 14, Extending Polyglot Notebooks, talks about how the Polyglot Notebooks experience can be customized and extended by providing custom formatters and magic commands.
Chapter 15, Adopting and Deploying Polyglot Notebooks, discusses ways of integrating Polyglot Notebooks into your workflows and sharing notebooks with others – including deploying them to a Jupyter Notebook server or hosting them online in a GitHub Codespace.
