TensorFlow.NET in Action

Name: TensorFlow.NET in Action
ISBN: 9787121443091

仇华 / 陈海平Publishing House of Electronics Industry · ChinaOct 2022

3.6

Word count: 135K words
Chapters: 32
ISBN: 9787121443091

Report an error Find this book

Summary

Built on the TensorFlow.NET framework, this book explains in detail the fundamental principles and applied techniques of deep learning on the .NET platform. It covers not only algorithm principles but also demonstrates practical code and running results, with complete examples written mainly in C# and F#.

The book is divided into three parts. Part 1 introduces core API usage and basic examples, including data types, tensors, Eager Mode, automatic differentiation, linear regression, logistic regression, tf.data, deep neural networks, and the AutoGraph mechanism, allowing readers to get started quickly. Part 2 focuses on demonstrating .NET Keras, including models, network layers, common APIs, model construction, and model training, so readers can master mainstream deep learning methods. Part 3 concentrates on production applications and hands-on case studies, including GPU environment setup, custom dataset training, image classification, object detection, transfer learning, natural language processing, generative adversarial networks, and F# application cases, each with complete code.

The book covers both the methods and practice of deep learning and is aimed primarily at university students, production technicians, and researchers. It is suitable for readers with a .NET programming foundation who wish to develop deep learning applications using .NET technologies.

Editor's review

Written for developers in the .NET ecosystem who want to move into deep learning, this book helps fill the shortage of Chinese-language practical material on TensorFlow.NET. It is organized progressively in three parts—core APIs, .NET Keras, and production cases—starting from foundational concepts such as tensors, automatic differentiation, and Eager Mode, moving on to model construction, training, and GPU environment deployment, and ending with complete cases in image classification, object detection, transfer learning, natural language processing, and generative adversarial networks, demonstrated in C# and F# with a clear hands-on orientation. Compared with pure-Python deep learning books, it places more emphasis on implementation within the .NET technology stack, making it suitable for readers who already have a programming foundation and need to embed deep learning capabilities into existing .NET projects step by step. The case code is complete and can serve as an engineering reference manual for on-demand consultation, but readers need to run and debug it themselves to turn API usage into real development ability.

Who it's for

A good fit for

Developers with a C# or F# foundation who want to do deep learning on the .NET platform
Production technicians who need to integrate deep learning models into existing .NET projects
University students who have studied machine learning fundamentals and want hands-on deep learning practice in C#
Researchers who need quick reference to TensorFlow.NET APIs and cases

Not a good fit for

Absolute beginners with no programming background and no exposure to .NET
Readers seeking systematic study of mathematical derivations and theoretical proofs in deep learning
Developers who work only in the Python ecosystem and do not intend to touch the .NET technology stack

Reading tips

Read Part 1 through first and run the example code section by section to build a solid foundation in tensors, automatic differentiation, and tf.data.
In Part 2, build the Keras models by hand chapter by chapter, and compare with the API documentation to understand each layer and training parameter.
In Part 3, read closely the cases closest to your own projects and use the remaining cases for reference.
Set up the GPU environment while reading; debugging as you go will significantly improve your understanding of production deployment.

About the publisher

Publishing House of Electronics Industry China · Founded in 1982

The Publishing House of Electronics Industry is a comprehensive publisher distinguished by its focus on electronic information and related fields. Judging from the more than 3,000 titles in its catalogue, its publishing covers programming and design, artificial intelligence, computer science, finance and economics, photography, and social sciences, attending both to cutting-edge technology and to general readers. Representative titles include Factor Investing: Methods and Practice, x86 Assembly Language: From Real Mode to Protected Mode, The RISC-V Reader, Artificial Intelligence: Speech Recognition, Understanding and Practice, Phased Array Radar Technology, Radar Target Characteristics, and 100 Go Mistakes and How to Avoid Them, most of which centre on technical methods and engineering practice. Its authors include Shi Chuan, Li Zhong, Wang Xiaobo, David Patterson, Andrew Waterman, and Yu Zihao, among others from China and abroad. Owing to limited public information, no further publishing features are elaborated here.

版权信息
内容简介
推荐序
序一
序二
第一部分 TensorFlow.NET API入门
第1章 TensorFlow.NET介绍
第2章数据类型与张量详解
第3章 Eager Mode详解
第4章自动求导原理与应用
第5章线性回归实操
第6章 MNIST手写字符分类逻辑回归
第7章 tf.data数据集创建与预处理
第8章深度神经网络实践
第9章 AutoGraph机制详解
第二部分 .NET Keras简明教程
第10章 Keras简要介绍
第11章模型与层
第12章 Keras常用API说明
第13章 Keras搭建模型的3种方式
第14章 Keras模型训练
第三部分生产应用与案例
第15章 CPU和GPU环境下的TensorFlow.NET应用
第16章工业生产环境应用案例
第17章在C#下使用TensorFlow.NET训练自己的数据集
第18章视觉图像分类
第19章视觉目标检测
第20章迁移学习应用
第21章自然语言处理
第22章生成对抗网络
第23章 F#应用案例
参考文献

FAQ

Can I read this book directly without a deep learning background?

You can get started, but you should first acquire some basic machine learning knowledge. The book begins with data types, tensors, Eager Mode, and other fundamentals, so Part 1 has a low barrier to entry; however, it emphasizes API usage and code practice and does not dwell on derivations such as backpropagation. Reading it alongside an introductory deep learning textbook will give more solid results.

What language is the example code mainly written in?

The complete examples are mainly written in C# and F#, and Part 3 also includes F# application cases. If you are familiar with C#, you can start on most content directly; the F# sections can be read as needed. Before reading, make sure the corresponding .NET development environment is installed on your machine and that you can run the examples in the book.

Can this book help me deploy models to a production environment?

Part 3 is specifically about production applications and includes complete cases such as GPU environment setup, custom dataset training, image classification, object detection, transfer learning, natural language processing, and generative adversarial networks, offering practical reference for pre-deployment engineering preparation. Actual deployment, however, still requires further practice based on the engineering conditions of your own system.

In what order should I read the three parts?

It is recommended to follow the book's order: use Part 1 to get started quickly with core APIs, use Part 2 to master model construction and training with .NET Keras, and then move on to production cases in Part 3. If you already have some foundation, you can skip some basic chapters, go directly to cases related to your own projects, and use the earlier chapters as a reference manual.

Comments

Find this book

This site is an index: you start “Search all sources” in your browser, and the entries it lists all point to third-party sites; please judge their availability yourself.

About the author

仇华 Author

Qiu Hua is an author about whom limited public information is available. Two works can currently be confirmed, both relating to the application of artificial intelligence technologies in the field of design. Their representative works, Technology and Practice of Green Prefabricated Revetment Structures in Water Network Areas and TensorFlow.NET in Action, address respectively revetment design in hydraulic engineering and the application of machine learning frameworks, spanning a fairly wide range of subjects but both oriented toward engineering practice and hands-on programming. Based on the available information, Qiu Hua is likely an author engaged in engineering technology or related technical development over the long term, and willing to organize practical experience into books. The works show a clear applied orientation and emphasize operability. Because public information is limited, personal details such as nationality and dates of birth and death cannot be confirmed. The two works fall under design and artificial intelligence respectively, which may suggest an interest in combining technical fields with real-world scenarios, though no excessive inference should be drawn.

陈海平 Author

Chen Haiping is an author about whom limited public information is available. Two works can currently be confirmed: TensorFlow.NET in Practice and The Forms and Colors of Materia Medica. These two books span notably different subjects, involving artificial intelligence and materia medica respectively, and are commonly classified under artificial intelligence and examination categories. TensorFlow.NET in Practice focuses on the practical application of the TensorFlow.NET framework, aimed at readers who wish to enter the field of machine learning through the .NET ecosystem, and is strongly hands-on in orientation. The Forms and Colors of Materia Medica relates to materia medica and reflects an approach centered on organizing and presenting traditional knowledge. Judging from these two works, the author handles both technical practice and traditional cultural subjects, with a writing style that emphasizes practicality and knowledge organization. As information on the author's life, nationality, and professional background is unavailable, a fuller picture of the author's work remains to be supplemented by further materials.