Use Java and Deeplearning4j to build robust, enterprise-grade deep learning models from scratch.
--- About This Book
* Install and configure Deeplearning4j and the TensorFlow Java API to implement deep learning models
* Explore recipes for training and fine-tuning your neural network models using Java
* Put your deep learning knowledge to use and train enterprise-grade neural networks with ease
-- Who This Book Is For
If you are a data scientist, machine learning developer, or a deep learning enthusiast who wants to implement deep learning models in Java, this book is for you. Basic understanding of Java programming as well as some experience with machine learning and neural networks is required to get the most out of this book.
--- What You Will Learn
* Perform data normalization and wrangling in Deeplearning4j
* Train, create, and evaluate deep learning models using DL4J
* Implement convolutional neural networks to solve image classification problems
* Train autoencoders and Generative Adversarial Networks (GANs) in Java
* Explore different ways to perform benchmarking and optimization
* Implement reinforcement learning for real-world use cases using RL4J
* Leverage the capability of DL4J in distributed systems
--- In Detail
Java is one of the most widely used programming languages in the world. With this book, you'll see how its popular libraries for deep learning, such as Deeplearning4j (DL4J), and the Java API for the TensorFlow package make deep learning easy. Starting by configuring DL4J to run on your GPU-powered machine, this deep learning cookbook will get you up to speed with troubleshooting installation issues. You'll then gain insights into deep learning basics and use your knowledge to create a deep neural network for binary classification from scratch. As you progress, you'll pick up on the technique of building a convolutional neural network (CNN) in DL4J, along with understanding how to construct numeric vectors from text. The book will also guide you through performing anomaly detection on unsupervised data and help you set up neural networks in distributed systems effectively. In addition to this, you'll learn to import models from Keras and change the configuration in a pre-trained DL4J model. Finally, you'll explore benchmarking in DL4J and optimize neural networks for optimal results. By the end of this book, you'll have a clear understanding of how you can use Deeplearning4j to build robust deep learning applications in Java.
Author(s): Rahul Raj
Edition: 1
Publisher: Packt
Year: 2020
Language: English
Pages: 294
Tags: machine learning java tensorflow Deeplearning4j
Title Page......Page 2
Copyright and Credits......Page 3
Dedication......Page 4
About Packt......Page 5
Contributors......Page 6
Table of Contents......Page 8
Preface......Page 16
Technical requirements......Page 22
Deep learning intuition......Page 23
Backpropagation......Page 24
Recurrent Neural Network (RNN)......Page 25
How to do it.........Page 26
How it works.........Page 27
There's more.........Page 30
How to do it.........Page 33
There's more.........Page 34
How it works.........Page 35
There's more.........Page 36
How it works.........Page 37
Getting ready......Page 39
How to do it.........Page 40
Getting ready......Page 41
How to do it.........Page 42
How it works.........Page 43
Troubleshooting installation issues......Page 44
How to do it.........Page 45
How it works.........Page 46
There's more.........Page 47
Chapter 2: Data Extraction, Transformation, and Loading......Page 49
Getting ready......Page 50
How to do it.........Page 51
How it works.........Page 55
How to do it.........Page 61
How it works.........Page 62
There's more.........Page 63
How to do it.........Page 64
There's more.........Page 65
How to do it.........Page 67
How to do it.........Page 68
How it works.........Page 69
There's more.........Page 70
How it works.........Page 71
There's more.........Page 72
Chapter 3: Building Deep Neural Networks for Binary Classification......Page 74
Extracting data from CSV input......Page 75
How it works.........Page 76
How to do it.........Page 77
How it works.........Page 78
There's more.........Page 80
How to do it.........Page 81
How it works.........Page 83
Getting ready......Page 85
Designing hidden layers for the neural network model......Page 86
Designing output layers for the neural network model......Page 87
How it works.........Page 88
How to do it.........Page 89
How it works.........Page 90
There's more.........Page 96
Getting ready......Page 98
How to do it.........Page 99
How it works.........Page 102
Chapter 4: Building Convolutional Neural Networks......Page 105
Technical requirements......Page 106
How it works.........Page 107
How to do it.........Page 109
How it works.........Page 110
How to do it.........Page 112
How it works.........Page 113
Constructing hidden layers for a CNN......Page 114
Constructing output layers for output classification......Page 115
Training images and evaluating CNN output......Page 116
How to do it.........Page 117
How it works.........Page 118
There's more.........Page 119
Creating an API endpoint for the image classifier......Page 120
How to do it.........Page 121
How it works.........Page 125
Chapter 5: Implementing Natural Language Processing......Page 127
Technical requirements......Page 128
Getting ready......Page 129
How to do it.........Page 130
There's more.........Page 132
How to do it.........Page 133
There's more.........Page 134
Evaluating the model......Page 135
How it works.........Page 136
How to do it.........Page 137
How it works.........Page 139
How it works.........Page 141
How it works.........Page 142
There's more.........Page 143
How to do it.........Page 144
How it works.........Page 145
Using Word2Vec for sentence classification using CNNs......Page 146
Getting ready......Page 147
How to do it.........Page 148
How it works.........Page 150
There's more.........Page 152
How to do it.........Page 153
How it works.........Page 155
Chapter 6: Constructing an LSTM Network for Time Series......Page 158
Technical requirements......Page 159
How to do it.........Page 160
How it works.........Page 161
Loading and transforming data......Page 162
How to do it.........Page 163
Constructing input layers for the network......Page 164
How it works.........Page 165
How to do it.........Page 166
How it works.........Page 167
How it works.........Page 168
How to do it.........Page 169
How it works.........Page 170
Chapter 7: Constructing an LSTM Neural Network for Sequence Classification......Page 172
Technical requirements......Page 173
How to do it.........Page 174
How it works.........Page 175
How to do it.........Page 177
How it works.........Page 178
How it works.........Page 180
How to do it.........Page 181
Constructing output layers for the network......Page 182
Evaluating the LSTM network for classified output......Page 183
How it works.........Page 184
Chapter 8: Performing Anomaly Detection on Unsupervised Data......Page 187
How to do it.........Page 188
How it works.........Page 189
Constructing dense layers for input......Page 190
Constructing output layers......Page 191
How it works.........Page 192
How it works.........Page 193
How to do it.........Page 194
How it works.........Page 195
How it works.........Page 197
There's more.........Page 198
Chapter 9: Using RL4J for Reinforcement Learning......Page 199
Technical requirements......Page 200
Setting up the Malmo environment and respective dependencies......Page 202
How to do it.........Page 203
Setting up the data requirements......Page 204
How to do it.........Page 205
How it works.........Page 208
How to do it.........Page 210
How it works.........Page 212
There's more.........Page 214
How to do it.........Page 215
How it works.........Page 216
Chapter 10: Developing Applications in a Distributed Environment......Page 218
Technical requirements......Page 219
How to do it.........Page 220
How it works.........Page 227
Creating an uber-JAR for training......Page 229
How to do it.........Page 230
How it works.........Page 231
How it works.........Page 232
There's more.........Page 233
How to do it.........Page 234
How it works.........Page 235
There's more.........Page 237
Configuring encoding thresholds......Page 238
How it works.........Page 239
There's more.........Page 240
How to do it.........Page 241
How it works..........Page 245
How to do it.........Page 246
Performing distributed inference......Page 247
How it works.........Page 248
Chapter 11: Applying Transfer Learning to Network Models......Page 249
Modifying an existing customer retention model......Page 250
How to do it.........Page 251
How it works.........Page 252
There's more.........Page 257
How to do it.........Page 258
Implementing frozen layers......Page 259
Importing and loading Keras models and layers......Page 260
How it works.........Page 261
Technical requirements......Page 264
Getting ready......Page 267
How to do it.........Page 268
How it works.........Page 269
There's more.........Page 271
How to do it.........Page 272
How it works.........Page 274
There's more.........Page 275
Using asynchronous ETL......Page 276
How it works.........Page 277
Using arbiter to monitor neural network behavior......Page 278
How it works.........Page 279
Performing hyperparameter tuning......Page 280
How to do it.........Page 281
How it works.........Page 284
Other Books You May Enjoy......Page 287
Index......Page 290
