TensorFlow Programming

Name: TensorFlow Programming
ISBN: 9787121486661

马斌 / 冯岭Publishing House of Electronics Industry · ChinaSep 2024

3.3

Word count: 86K words
Chapters: 85
ISBN: 9787121486661

Report an error Find this book

Summary

This book provides a comprehensive introduction to the TensorFlow 2.x framework and its applications in deep learning, covering an overview of TensorFlow, Python fundamentals, environment setup and getting started, TensorBoard visualization, implementing multilayer perceptrons, convolutional neural networks, and recurrent neural networks, reinforcement learning, transfer learning, generative adversarial networks, and GPU parallel computing.

Editor's review

Centered on TensorFlow 2.x, this book covers both framework fundamentals and the main deep learning models in a single volume. It first lays out Python basics and environment setup, then moves progressively into TensorBoard visualization, multilayer perceptrons, convolutional neural networks, and recurrent neural networks, before extending to reinforcement learning, transfer learning, generative adversarial networks, and GPU parallel computing. This structure is beginner-friendly: it helps readers build a complete picture of the workflow from environment configuration to model implementation, and makes it easy to compare different network architectures within the same framework. The coverage is broad, making the book suitable as a main reference for coursework or for a self-study path, especially for readers who want hands-on experience running examples and understanding the common TensorFlow 2.x APIs. When reading, it is best not to just look at the code but to modify datasets, tune parameters, and observe visualization results for each chapter's topic, so as to connect framework usage with model ideas. For readers already familiar with deep learning theory who only want to look up specific APIs, the introductory groundwork may feel excessive; the book is better suited to systematic study than to quick reference.

Who it's for

A good fit for

Undergraduate or graduate students with Python basics who want a systematic introduction to TensorFlow 2.x
Beginners and career changers who need to turn deep learning models into working code
Instructors teaching deep learning courses who need a reference covering multiple network architectures
Readers who want to learn how reinforcement learning, transfer learning, and GANs are implemented in TensorFlow

Not a good fit for

Readers with no programming background who cannot yet read Python code
Experienced developers seeking a direct reference for advanced TensorFlow API details
Readers focused solely on cutting-edge research in one specific area who do not need systematic introductory content

Reading tips

Set up the environment and run the examples chapter by chapter first, then return to understand the design rationale of each chapter's model.
When reaching convolutional and recurrent neural networks, swap in your own datasets and observe how the training curves change.
Use TensorBoard visualization as a debugging tool, recording at least one experiment comparison per chapter.
For reinforcement learning, transfer learning, and GANs, first understand the workflow, then dive deeper into one chapter of interest.

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
编写组
总序
前言
第1章 TensorFlow简介
1.1 人工智能的编程框架
1.2 TensorFlow 与人工智能
1.3 TensorFlow 数据模型
1.4 TensorFlow 计算模型和运行模型
1.5 实验：矩阵运算
习题
第2章 Python语言基础
2.1 Python语言
2.2 基础语法
2.3 数据结构
2.4 面向对象特性
2.5 其他高级特性
2.6 实验：Python基本语法的实现
习题
第3章环境搭建与入门
3.1 开发平台简介
3.2 开发环境部署
3.3 一个简单的实例
习题
第4章 TensorBoard可视化
4.1 什么是TensorBoard
4.2 基本流程与结构
4.3 图表的可视化
4.4 监控指标的可视化
4.5 学习过程的可视化
4.6 实验：TensorBoard可视化实现
习题
第5章多层感知机实现
5.1 感知机
5.2 多层感知机与前向传播
5.3 前向传播
5.4 梯度下降
5.5 反向传播
5.6 数据集
5.7 多层感知机的实现
5.8 实验：基于Keras多层感知机的MNIST手写数字识别
习题
第6章卷积神经网络实现
6.1 CNN基本原理
6.2 CNN的卷积操作
6.3 CNN的池化操作
6.4 使用简单的CNN实现手写字符识别
6.5 AlexNet
6.6 实验：基于VGG16模型的图像分类实现
习题
第7章循环神经网络实现
7.1 RNN简介
7.2 长短时记忆网络
7.3 双向RNN
7.4 深层RNN
7.5 实验：基于LSTM的股票预测
习题
第8章强化学习
8.1 强化学习原理
8.2 马尔可夫决策过程实现
8.3 基于价值的强化学习方法
8.4 Gym的简单使用
8.5 实验：基于强化学习的小车爬山游戏
习题
第9章迁移学习
9.1 迁移学习原理
9.2 基于模型的迁移学习方法实现
9.3 基于VGG-19的迁移学习实现
9.4 实验：基于Inception V3的迁移学习
习题
第10章生成对抗网络
10.1 GAN概述
10.2 GAN的目标函数
10.3 GAN的实现
10.4 深度卷积生成对抗网络
10.5 GAN的衍生模型
习题
第11章 GPU并行计算
11.1 并行计算技术
11.2 TensorFlow加速方法
11.3 单GPU并行加速的实现
11.4 多GPU并行加速的实现
11.5 实验：基于GPU的矩阵乘法
习题

FAQ

Can I read this book without a deep learning background?

Yes, but you need Python basics first. The book starts from environment setup and getting started, then gradually moves into multilayer perceptrons, convolutional neural networks, and more, so it is suitable for learning by doing. If you have no understanding of basic neural network concepts at all, it is better to pick up some introductory knowledge first for a better experience.

Which TensorFlow version does the book's content correspond to?

According to the available information, the book introduces the TensorFlow 2.x framework and its applications in deep learning, covering environment setup and getting started, TensorBoard visualization, and implementations of various network architectures. When reading, it is recommended to configure and run the examples in a TensorFlow 2.x environment.

Is this book suitable as a course textbook?

The book covers everything from framework fundamentals to multilayer perceptrons, convolutional neural networks, and recurrent neural networks, and further to reinforcement learning, transfer learning, and generative adversarial networks. Its structure is fairly complete, making it suitable as a companion textbook for deep learning or TensorFlow-related courses, and convenient for instructors to organize experiments chapter by chapter.

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

马斌 Author

Ma Bin's publicly available information is currently limited, and no systematic account of his identity or career has yet appeared. Still, the three works recorded in the library suggest a fairly wide range of interests. He has written a technical book such as TensorFlow Programming, aimed at programming and artificial-intelligence practice; he has also contributed to a basic tutorial on the Chinese edition of AutoCAD 2022, which focuses on getting started with engineering drawing software; and in Learn Communication in One Book: These Are the Key Skills, he turns to interpersonal communication and management. Spanning programming and design as well as management, his works are practical and introductory in character, concentrating on explaining specific tools and methods clearly so that beginners can follow them step by step. Because public information about his life and professional background is lacking, his writing style and influence can only be summarized as above from the available titles, and further details await additional sources.

冯岭 Author

Feng Ling is an author about whom little public information is available; to date, only one work in the field of programming and design is known, TensorFlow Programming. Judging from its subject matter, Feng Ling focuses on the engineering applications of the deep learning framework TensorFlow, addressing readers who wish to use this tool for model building and program implementation. Writing of this kind generally requires a clear grasp of the framework's interfaces, computational graph mechanisms, and common development workflows, as well as the ability to organize technical points into chapters that are easy to study. Because only this one work is included in the collection, Feng Ling's other writing directions, professional background, and career experience cannot be confirmed, and the work's actual influence awaits further information. On the available evidence, Feng Ling may be understood as a Chinese author engaged in technical writing and knowledge organization around a mainstream deep learning framework, whose work offers Chinese readers an introductory path to learning TensorFlow programming.