Java Deep Learning Projects : Implement 10 Real-world Deep Learning Applications Using Deeplearning4j and Open Source APIs

Build and deploy powerful neural network models using the latest Java deep learning librariesKey FeaturesUnderstand DL with Java by implementing real-world projectsMaster implementations of various ANN models and build your own DL systemsDevelop applications using NLP, image classification, RL, and GPU processingBook DescriptionJava is one of the most widely used programming languages. With the rise of deep learning, it has become a popular choice of tool among data scientists and machine learning experts.Java Deep Learning Projects starts with an overview of deep learning concepts and then delves into advanced projects. You will see how to build several projects using different deep neural network architectures such as multilayer perceptrons, Deep Belief Networks, CNN, LSTM, and Factorization Machines.You will get acquainted with popular deep and machine learning libraries for Java such as Deeplearning4j, Spark ML, and RankSys and you'll be able to use their features to build and deploy projects on distributed computing environments.You will then explore advanced domains such as transfer learning and deep reinforcement learning using the Java ecosystem, covering various real-world domains such as healthcare, NLP, image classification, and multimedia analytics with an easy-to-follow approach. Expert reviews and tips will follow every project to give you insights and hacks.By the end of this book, you will have stepped up your expertise when it comes to deep learning in Java, taking it beyond theory and be able to build your own advanced deep learning systems.What you will learnMaster deep learning and neural network architecturesBuild real-life applications covering image classification, object detection, online trading, transfer learning, and multimedia analytics using DL4J and open-source APIsTrain ML agents to learn from data using deep reinforcement learningUse factorization machines for advanced movie recommendationsTrain DL models on distributed GPUs for faster deep learning with Spark and DL4JEase your learning experience through 69 FAQsWho This Book Is ForIf you are a data scientist, machine learning professional, or deep learning practitioner keen to expand your knowledge by delving into the practical aspects of deep learning with Java, then this book is what you need! Get ready to build advanced deep learning models to carry out complex numerical computations. Some basic understanding of machine learning concepts and a working knowledge of Java are required.Table of ContentsGetting Started with Deep LearningCancer Type Prediction using Recurrent Type NetworksImage Classification using Convolutional Neural NetworksSentiment Analysis using Word2Vec and LSTM NetworksImage Classification using Transfer LearningReal-Time Object Detection Using YOLO, JavaCV, and DL4JStock Price Prediction Using the LSTM NetworkDistributed Deep Learning – Video Classification Using Convolutional-LSTM NetworksUsing Deep Reinforcement Learning for a GridWorld GameMovie Recommendation System using Factorization MachinesDiscussion, Current Trends, and Outlook Title Page......Page 2 Copyright and Credits......Page 3 Java Deep Learning Projects......Page 4 Packt Upsell......Page 5 Why subscribe?......Page 6 PacktPub.com......Page 7 Contributors......Page 8 About the author......Page 9 About the reviewer......Page 10 Packt is searching for authors like you......Page 11 Preface......Page 25 Who this book is for......Page 26 What this book covers......Page 27 To get the most out of this book......Page 29 Download the example code files......Page 31 Download the color images......Page 32 Conventions used......Page 33 Get in touch......Page 34 Reviews......Page 35 Getting Started with Deep Learning......Page 36 A soft introduction to ML......Page 37 Working principles of ML algorithms......Page 38 Supervised learning......Page 41 Unsupervised learning......Page 43 Reinforcement learning......Page 45 Putting ML tasks altogether......Page 46 Delving into deep learning......Page 47 How did DL take ML into next level?......Page 48 Artificial Neural Networks......Page 52 Biological neurons......Page 53 A brief history of ANNs......Page 55 How does an ANN learn?......Page 57 ANNs and the backpropagation algorithm......Page 58 Forward and backward passes......Page 59 Weights and biases......Page 60 Weight optimization......Page 63 Activation functions......Page 65 Neural network architectures......Page 67 Deep neural networks......Page 68 Multilayer Perceptron......Page 70 Deep belief networks......Page 71 Autoencoders......Page 73 Convolutional neural networks......Page 74 Recurrent neural networks ......Page 76 Emergent architectures......Page 77 Residual neural networks......Page 78 Generative adversarial networks......Page 79 Capsule networks......Page 80 DL frameworks and cloud platforms......Page 81 Deep learning frameworks......Page 82 Cloud-based platforms for DL......Page 85 Deep learning from a disaster – Titanic survival prediction......Page 86 Problem description......Page 87 Configuring the programming environment......Page 90 Feature engineering and input dataset preparation......Page 92 Training MLP classifier ......Page 99 Evaluating the MLP classifier......Page 101 Frequently asked questions (FAQs)......Page 105 Summary......Page 106 Answers to FAQs......Page 107 Cancer Types Prediction Using Recurrent Type Networks......Page 113 Deep learning in cancer genomics......Page 114 Cancer genomics dataset description......Page 117 Preparing programming environment......Page 123 Titanic survival revisited with DL4J......Page 126 Multilayer perceptron network construction......Page 129 Hidden layer 1......Page 131 Hidden layer 2......Page 132 Output layer......Page 133 Network training......Page 135 Evaluating the model......Page 137 Cancer type prediction using an LSTM network......Page 140 Dataset preparation for training......Page 141 Recurrent and LSTM networks......Page 145 Dataset preparation......Page 150 LSTM network construction......Page 151 Network training......Page 153 Evaluating the model......Page 154 Frequently asked questions (FAQs)......Page 156 Summary......Page 157 Answers to questions......Page 158 Multi-Label Image Classification Using Convolutional Neural Networks......Page 166 Image classification and drawbacks of DNNs......Page 167 CNN architecture......Page 169 Convolutional operations......Page 171 Pooling and padding operations......Page 172 Fully connected layer (dense layer)......Page 175 Multi-label image classification using CNNs......Page 176 Problem description......Page 177 Description of the dataset......Page 179 Removing invalid images......Page 181 Workflow of the overall project......Page 182 Image preprocessing......Page 185 Extracting image metadata......Page 191 Image feature extraction......Page 193 Preparing the ND4J dataset......Page 200 Training, evaluating, and saving the trained CNN models......Page 202 Network construction......Page 203 Scoring the model......Page 208 Submission file generation......Page 209 Wrapping everything up by executing the main() method......Page 211 Frequently asked questions (FAQs)......Page 213 Summary......Page 214 Answers to questions......Page 215 Sentiment Analysis Using Word2Vec and LSTM Network......Page 221 Sentiment analysis is a challenging task......Page 223 Using Word2Vec for neural word embeddings......Page 227 Datasets and pre-trained model description......Page 229 Large Movie Review dataset for training and testing......Page 230 Folder structure of the dataset......Page 231 Description of the sentiment labeled dataset......Page 233 Word2Vec pre-trained model......Page 234 Sentiment analysis using Word2Vec and LSTM......Page 235 Preparing the train and test set using the Word2Vec model......Page 236 Network construction, training, and saving the model......Page 244 Restoring the trained model and evaluating it on the test set......Page 247 Making predictions on sample review texts......Page 249 Frequently asked questions (FAQs)......Page 253 Summary......Page 254 Answers to questions......Page 255 Transfer Learning for Image Classification......Page 263 Image classification with pretrained VGG16......Page 264 DL4J and transfer learning......Page 266 Developing an image classifier using transfer learning......Page 267 Dataset collection and description......Page 268 Architecture choice and adoption......Page 269 Train and test set preparation......Page 274 Network training and evaluation......Page 276 Restoring the trained model and inferencing......Page 279 Making simple inferencing......Page 281 Frequently asked questions (FAQs)......Page 284 Summary......Page 285 Answers to questions......Page 286 Real-Time Object Detection using YOLO, JavaCV, and DL4J......Page 288 Object detection from images and videos......Page 289 Object classification, localization, and detection......Page 290 Convolutional Sliding Window (CSW)......Page 294 Object detection from videos......Page 298 You Only Look Once (YOLO)......Page 300 Developing a real-time object detection project......Page 305 Step 1 – Loading a pre-trained YOLO model......Page 306 Step 2 – Generating frames from video clips......Page 309 Step 3 – Feeding generated frames into Tiny YOLO model......Page 312 Step 4 – Object detection from image frames......Page 314 Step 5 – Non-max suppression in case of more than one bounding box......Page 316 Step 6 – wrapping up everything and running the application......Page 319 Frequently asked questions (FAQs)......Page 323 Summary......Page 324 Answers to questions......Page 325 Stock Price Prediction Using LSTM Network......Page 326 State-of-the-art automated stock trading......Page 327 Developing a stock price predictive model......Page 331 Data collection and exploratory analysis......Page 333 Preparing the training and test sets......Page 339 LSTM network construction......Page 347 Network training, and saving the trained model......Page 350 Restoring the saved model for inferencing......Page 353 Evaluating the model......Page 354 Frequently asked questions (FAQs)......Page 362 Summary......Page 363 Answers to questions......Page 364 Distributed Deep Learning – Video Classification Using Convolutional LSTM Networks......Page 366 Distributed deep learning across multiple GPUs......Page 367 Distributed training on GPUs with DL4J......Page 369 Video classification using convolutional – LSTM......Page 372 UCF101 – action recognition dataset......Page 373 Preprocessing and feature engineering......Page 375 Solving the encoding problem......Page 376 Data processing workflow......Page 378 Simple UI for checking video frames......Page 383 Preparing training and test sets......Page 385 Network creation and training......Page 387 Performance evaluation......Page 391 Distributed training on AWS deep learning AMI 9.0......Page 393 Frequently asked questions (FAQs)......Page 403 Summary......Page 404 Answers to questions......Page 405 Playing GridWorld Game Using Deep Reinforcement Learning......Page 407 Notation, policy, and utility for RL......Page 408 Notations in reinforcement learning......Page 409 Policy......Page 411 Utility......Page 412 Neural Q-learning......Page 413 Introduction to QLearning......Page 414 Neural networks as a Q-function......Page 416 Developing a GridWorld game using a deep Q-network......Page 419 Generating the grid......Page 421 Calculating agent and goal positions......Page 423 Calculating the action mask......Page 425 Providing guidance action......Page 426 Calculating the reward......Page 428 Flattening input for the input layer......Page 429 Network construction and training......Page 430 Playing the GridWorld game......Page 438 Frequently asked questions (FAQs)......Page 441 Summary......Page 442 Answers to questions......Page 443 Developing Movie Recommendation Systems Using Factorization Machines......Page 447 Recommendation systems......Page 448 Recommendation approaches......Page 449 Collaborative filtering approaches......Page 450 Content-based filtering approaches......Page 452 Hybrid recommender systems......Page 453 Model-based collaborative filtering......Page 454 The utility matrix......Page 455 The cold-start problem in collaborative-filtering approaches......Page 457 Factorization machines in recommender systems......Page 458 Developing a movie recommender system using FMs......Page 461 Dataset description and exploratory analysis......Page 463 Movie rating prediction......Page 470 Converting the dataset into LibFM format......Page 471 Training and test set preparation......Page 476 Movie rating prediction......Page 479 Which one makes more sense ;– ranking or rating?......Page 491 Frequently asked questions (FAQs)......Page 501 Summary......Page 502 Answers to questions......Page 503 Discussion, Current Trends, and Outlook......Page 507 Discussion and outlook......Page 508 Discussion on the completed projects......Page 509 Titanic survival prediction using MLP and LSTM networks......Page 510 Cancer type prediction using recurrent type networks......Page 511 Image classification using convolutional neural networks......Page 512 Sentiment analysis using Word2Vec and the LSTM network......Page 513 Image classification using transfer learning......Page 514 Real-time object detection using YOLO, JavaCV, and DL4J......Page 515 Stock price prediction using LSTM network......Page 516 Distributed deep learning – video classification using a convolutional-LSTM network......Page 517 Using deep reinforcement learning for GridWorld......Page 518 Movie recommender system using factorization machines......Page 519 Current trends and outlook......Page 520 Current trends......Page 521 Outlook on emergent DL architectures......Page 522 Residual neural networks......Page 523 GANs......Page 524 Capsule networks (CapsNet)......Page 525 Semantic image segmentation......Page 527 Deep learning for clustering analysis......Page 528 Frequently asked questions (FAQs)......Page 530 Answers to questions......Page 531 Other Books You May Enjoy......Page 533 Leave a review - let other readers know what you think......Page 535 Annotation Build and deploy powerful neural network models using the latest Java deep learning librariesKey Features Understand DL with Java by implementing real-world projects Master implementations of various ANN models and build your own DL systems Develop applications using NLP, image classification, RL, and GPU processingBook DescriptionJava is one of the most widely used programming languages. With the rise of deep learning, it has become a popular choice of tool among data scientists and machine learning experts. Java Deep Learning Projects starts with an overview of deep learning concepts and then delves into advanced projects. You will see how to build several projects using different deep neural network architectures such as multilayer perceptrons, Deep Belief Networks, CNN, LSTM, and Factorization Machines. You will get acquainted with popular deep and machine learning libraries for Java such as Deeplearning4j, Spark ML, and RankSys and you'll be able to use their features to build and deploy projects on distributed computing environments. You will then explore advanced domains such as transfer learning and deep reinforcement learning using the Java ecosystem, covering various real-world domains such as healthcare, NLP, image classification, and multimedia analytics with an easy-to-follow approach. Expert reviews and tips will follow every project to give you insights and hacks. By the end of this book, you will have stepped up your expertise when it comes to deep learning in Java, taking it beyond theory and be able to build your own advanced deep learning systems. What you will learnMaster deep learning and neural network architecturesBuild real-life applications covering image classification, object detection, online trading, transfer learning, and multimedia analytics using DL4J and open-source APIsTrain ML agents to learn from data using deep reinforcement learningUse factorization machines for advanced movie recommendationsTrain DL models on distributed GPUs for faster deep learning with Spark and DL4JEase your learning experience through 69 FAQsWho this book is forIf you are a data scientist, machine learning professional, or deep learning practitioner keen to expand your knowledge by delving into the practical aspects of deep learning with Java, then this book is what you need! Get ready to build advanced deep learning models to carry out complex numerical computations. Some basic understanding of machine learning concepts and a working knowledge of Java are required You will build full-fledged, deep learning applications with Java and different open-source libraries. Master numerical computing, deep learning, and the latest Java programming features to carry out complex advanced tasks. This book is filled with best practices/tips after every project to help you optimize your deep learning models with ease.