یادگیری تنسورفلو ۲.۰: پیاده‌سازی مدل‌های یادگیری ماشین و یادگیری عمیق با پایتون

Learn how to use TensorFlow 2.0 to build machine learning and deep learning models with complete examples. The book begins with introducing TensorFlow 2.0 framework and the major changes from its last release. Next, it focuses on building Supervised Machine Learning models using TensorFlow 2.0. It also demonstrates how to build models using customer estimators. Further, it explains how to use TensorFlow 2.0 API to build machine learning and deep learning models for image classification using the standard as well as custom parameters. You'll review sequence predictions, saving, serving, deploying, and standardized datasets, and then deploy these models to production. All the code presented in the book will be available in the form of executable scripts at Github which allows you to try out the examples and extend them in interesting ways. What You'll Learn: • Review the new features of TensorFlow 2.0 • Use TensorFlow 2.0 to build machine learning and deep learning models • Perform sequence predictions using TensorFlow 2.0 • Deploy TensorFlow 2.0 models with practical examples Who This Book Is For: Data scientists, machine and deep learning engineers. Table of Contents 5 About the Authors 9 About the Technical Reviewer 11 Acknowledgments 12 Introduction 14 Chapter 1: Introduction to TensorFlow 2.0 16 Tensor + Flow = TensorFlow 17 Components and Basis Vectors 18 Tensor 21 Rank 22 Shape 22 Flow 22 TensorFlow 1.0 vs. TensorFlow 2.0 24 Usability-Related Changes 25 Simpler APIs 25 High-Level APIs 25 Lower-Level APIs 26 Session Execution 26 Eager Execution 27 tf.function 29 Keras 29 Redundancy 30 Improved Documentation and More Inbuilt Data Sources 31 Performance-Related Changes 31 Installation and Basic Operations in TensorFlow 2.0 32 Anaconda 32 Colab 32 Databricks 34 Conclusion 39 Chapter 2: Supervised Learning with TensorFlow 40 What Is Supervised Machine Learning? 40 Linear Regression with TensorFlow 2.0 43 Implementation of a Linear Regression Model, Using TensorFlow and Keras 44 Logistic Regression with TensorFlow 2.0 52 Boosted Trees with TensorFlow 2.0 62 Ensemble Technique 62 Bagging 63 Boosting 63 Gradient Boosting 64 Conclusion 67 Chapter 3: Neural Networks and Deep Learning with TensorFlow 68 What Are Neural Networks? 68 Neurons 69 Artificial Neural Networks (ANNs) 70 Simple Neural Network Architecture 72 Forward and Backward Propagation 73 Building Neural Networks with TensorFlow 2.0 76 About the Data Set 76 Deep Neural Networks (DNNs) 82 Building DNNs with TensorFlow 2.0 83 Estimators Using the Keras Model 86 Conclusion 89 Chapter 4: Images with TensorFlow 90 Image Processing 91 Convolutional Neural Networks 92 Convolutional Layer 92 Pooling Layer 95 Fully Connected Layer 96 ConvNets Using TensorFlow 2.0 97 Advanced Convolutional Neural Network Architectures 104 Transfer Learning 108 Transfer Learning and Machine Learning 110 Variational Autoencoders Using TensorFlow 2.0 112 Autoencoders 112 Applications of Autoencoders 113 Variational Autoencoders 113 Implementation of Variational Autoencoders Using TensorFlow 2.0 114 Conclusion 121 Chapter 5: Natural Language Processing with TensorFlow 2.0 122 NLP Overview 122 Text Preprocessing 124 Tokenization 125 Word Embeddings 127 Text Classification Using TensorFlow 128 Text Processing 130 Deep Learning Model 134 Embeddings 135 TensorFlow Projector 138 Conclusion 144 Chapter 6: TensorFlow Models in Production 145 Model Deployment 146 Isolation 147 Collaboration 147 Model Updates 148 Model Performance 148 Load Balancer 148 Python-Based Model Deployment 149 Saving and Restoring a Machine Learning Model 149 Deploying a Machine Learning Model As a REST Service 152 Templates 156 Challenges of Using Flask 159 Building a Keras TensorFlow-Based Model 160 TF ind deployment 165 Conclusion 173 Index 174 Learn how to use TensorFlow 2.0 to build machine learning and deep learning models with complete examples. The book begins with introducing TensorFlow 2.0 framework and the major changes from its last release. Next, it focuses on building Supervised Machine Learning models using TensorFlow 2.0. It also demonstrates how to build models using customer estimators. Further, it explains how to use TensorFlow 2.0 API to build machine learning and deep learning models for image classification using the standard as well as custom parameters. You'll review sequence predictions, saving, serving, deploying, and standardized datasets, and then deploy these models to production. All the code presented in the book will be available in the form of executable scripts at Github which allows you to try out the examples and extend them in interesting ways. You will: Review the new features of TensorFlow 2.0 Use TensorFlow 2.0 to build machine learning and deep learning models Perform sequence predictions using TensorFlow 2.0 Deploy TensorFlow 2.0 models with practical examples Front Matter ....Pages i-xvi Introduction to TensorFlow 2.0 (Pramod Singh, Avinash Manure)....Pages 1-24 Supervised Learning with TensorFlow (Pramod Singh, Avinash Manure)....Pages 25-52 Neural Networks and Deep Learning with TensorFlow (Pramod Singh, Avinash Manure)....Pages 53-74 Images with TensorFlow (Pramod Singh, Avinash Manure)....Pages 75-106 Natural Language Processing with TensorFlow 2.0 (Pramod Singh, Avinash Manure)....Pages 107-129 TensorFlow Models in Production (Pramod Singh, Avinash Manure)....Pages 131-159 Back Matter ....Pages 161-164