Deep learning with tensorflow explore neural networks and build intelligent systems with python 2nd edition fully revised and updated

Table of Contents Deep Learning with TensorFlow - Second Edition Why subscribe? PacktPub.com Contributors About the authors About the reviewers Packt is Searching for Authors Like You Preface Who this book is for What this book covers To get the most out of this book Download the example code files Download the color images Conventions used Get in touch Reviews Getting Started with Deep Learning A soft introduction to machine learning Supervised learning Unbalanced data Unsupervised learning Reinforcement learning What is deep learning? Artificial neural networks The biological neurons The artificial neuron How does an ANN learn? ANNs and the backpropagation algorithm Weight optimization Stochastic gradient descent Neural network architectures Deep Neural Networks (DNNs) Multilayer perceptron Deep Belief Networks (DBNs) Convolutional Neural Networks (CNNs) AutoEncoders Recurrent Neural Networks (RNNs) Emergent architectures Deep learning frameworks Summary A First Look at TensorFlow A general overview of TensorFlow What's new from TensorFlow v1.6 forwards? Nvidia GPU support optimized Introducing TensorFlow Lite Eager execution Optimized Accelerated Linear Algebra (XLA) Installing and configuring TensorFlow TensorFlow computational graph TensorFlow code structure Eager execution with TensorFlow Data model in TensorFlow Tensor Rank and shape Data type Variables Fetches Feeds and placeholders Visualizing computations through TensorBoard How does TensorBoard work? Linear regression and beyond Linear regression revisited for a real dataset Summary Feed-Forward Neural Networks with TensorFlow Feed-forward neural networks (FFNNs) Feed-forward and backpropagation Weights and biases Activation functions Using sigmoid Using Using ReLU Using softmax Implementing a feed-forward neural network Exploring the MNIST dataset Softmax classifier Implementing a multilayer perceptron (MLP) Training an MLP Using MLPs Dataset description Preprocessing A TensorFlow implementation of MLP for client-subscription assessment Deep Belief Networks (DBNs) Restricted Boltzmann Machines (RBMs) Construction of a simple DBN Unsupervised pre-training Supervised fine-tuning Implementing a DBN with TensorFlow for client-subscription assessment Tuning hyperparameters and advanced FFNNs Tuning FFNN hyperparameters Number of hidden layers Number of neurons per hidden layer Weight and biases initialization Selecting the most suitable optimizer GridSearch and randomized search for hyperparameters tuning Regularization Dropout optimization Summary Convolutional Neural Networks Main concepts of CNNs CNNs in action LeNet5 Implementing a LeNet-5 step by step AlexNet Transfer learning Pretrained AlexNet Dataset preparation Fine-tuning implementation VGG Artistic style learning with VGG-19 Input images Content extractor and loss Style extractor and loss Merger and total loss Training Inception-v3 Exploring Inception with TensorFlow Emotion recognition with CNNs Testing the model on your own image Source code Summary Optimizing TensorFlow Autoencoders How does an autoencoder work? Implementing autoencoders with TensorFlow Improving autoencoder robustness Implementing a denoising autoencoder Implementing a convolutional autoencoder Encoder Decoder Fraud analytics with autoencoders Description of the dataset Problem description Exploratory data analysis Training, validation, and testing set preparation Normalization Autoencoder as an unsupervised feature learning algorithm Evaluating the model Summary Recurrent Neural Networks Working principles of RNNs Implementing basic RNNs in TensorFlow RNN and the long-term dependency problem Bi-directional RNNs RNN and the gradient vanishing-exploding problem LSTM networks GRU cell Implementing an RNN for spam prediction Data description and preprocessing Developing a predictive model for time series data Description of the dataset Pre-processing and exploratory analysis LSTM predictive model Model evaluation An LSTM predictive model for sentiment analysis Network design LSTM model training Visualizing through TensorBoard LSTM model evaluation Human activity recognition using LSTM model Dataset description Workflow of the LSTM model for HAR Implementing an LSTM model for HAR Summary Heterogeneous and Distributed Computing GPGPU computing The GPGPU history The CUDA architecture The GPU programming model The TensorFlow GPU setup Update TensorFlow GPU representation Using a GPU GPU memory management Assigning a single GPU on a multi-GPU system The source code for GPU with soft placement Using multiple GPUs Distributed computing Model parallelism Data parallelism The distributed TensorFlow setup Summary Advanced TensorFlow Programming tf.estimator Estimators Graph actions Parsing resources Flower predictions TFLearn Installation Titanic survival predictor PrettyTensor Chaining layers Normal mode Sequential mode Branch and join Digit classifier Keras Keras programming models Sequential model Sentiment classification of movie reviews Functional API SqueezeNet Summary Recommendation Systems Using Factorization Machines Recommendation systems Collaborative filtering approaches Content-based filtering approaches Hybrid recommender systems Model-based collaborative filtering Movie recommendation using collaborative filtering The utility matrix Description of the dataset Ratings data Movies data Users data Exploratory analysis of the MovieLens dataset Implementing a movie RE Training the model with the available ratings Inferencing the saved model Generating the user-item table Clustering similar movies Movie rating prediction by users Finding top k movies Predicting top k similar movies Computing user-user similarity Evaluating the recommender system Factorization machines for recommendation systems Factorization machines Cold-start problem and collaborative-filtering approaches Problem definition and formulation Dataset description Workflow of the implementation Preprocessing Training the FM model Improved factorization machines Neural factorization machines Dataset description Using NFM for the movie recommendation Model training Model evaluation Summary 10 Reinforcement Learning The RL problem OpenAI Gym OpenAI environments The env class Installing and running OpenAI Gym The Q-Learning algorithm The FrozenLake environment Deep Q-learning Deep Q neural networks The Cart-Pole problem Deep Q-Network for the Cart-Pole problem The Experience Replay method Exploitation and exploration The Deep Q-Learning training algorithm Summary Other Books You May Enjoy Leave a review – let other readers know what you think Index Deep Learning with TensorFlow Second Edition Deep Learning with TensorFlow Second Edition Copyright © 2018 Packt Publishing All rights reserved No part of this book may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, without the prior written permission of the publisher, except in the case of brief quotations embedded in critical articles or reviews Every effort has been made in the preparation of this book to ensure the accuracy of the information presented However, the information contained in this book is sold without warranty, either express or implied Neither the authors, nor Packt Publishing or its dealers and distributors, will be held liable for any damages caused or alleged to have been caused directly or indirectly by this book Packt Publishing has endeavored to provide trademark information about all of the companies and products mentioned in this book by the appropriate use of capitals However, Packt Publishing cannot guarantee the accuracy of this information Acquisition Editors: Ben Renow-Clarke, Suresh Jain Project Editor: Savvy Sequeira Content Development Editors: Jo Lovell Technical Editor: Nidhisha Shetty Copy Editor: Safis Editing Indexers: Tejal Daruwale Soni Graphics: Tom Scaria Production Coordinator: Arvindkumar Gupta First published: April 2017 Second edition: March 2018 Production reference: 1290318 Published by Packt Publishing Ltd Livery Place 35 Livery Street Birmingham B3 2PB, UK ISBN 978-1-78883-110-9 www.packtpub.com mapt.io Mapt is an online digital library that gives you full access to over 5,000 books and videos, as well as industry leading tools to help you plan your personal development and advance your career For more information, please visit our website Why subscribe? Spend less time learning and more time coding with practical eBooks and Videos from over 4,000 industry professionals Learn better with Skill Plans built especially for you Get a free eBook or video every month Mapt is fully searchable Copy and paste, print, and bookmark content GoogLeNet about / Inception-v3 GPGPU about / The GPGPU history GPGPU computing about / GPGPU computing GPGPU history / The GPGPU history CUDA architecture / The CUDA architecture GPU programming model / The GPU programming model GPU programming model working / The GPU programming model Gradient Descent (GD) about / Weight optimization Gram matrix / Style extractor and loss greedy policy about / Exploitation and exploration GridSearchCV reference / GridSearch and randomized search for hyperparameters tuning H Human Activity Recognition (HAR), with LSTM model about / Human activity recognition using LSTM model dataset description / Dataset description workflow / Workflow of the LSTM model for HAR implementation / Implementing an LSTM model for HAR hybrid recommender systems about / Hybrid recommender systems hyperparameters about / A soft introduction to machine learning I improved factorization machines about / Improved factorization machines Neural Factorization Machines (NFMs) / Neural factorization machines Inception-v3 about / Inception-v3 exploring, with TensorFlow / Exploring Inception with TensorFlow Inception vN about / Inception-v3 input neurons / Implementing a multilayer perceptron (MLP) IPython Notebook reference / Data type issues, collaborative filtering approaches cold start / Collaborative filtering approaches scalability / Collaborative filtering approaches sparsity / Collaborative filtering approaches K K-means / What is deep learning? Kaggle platform URL / Emotion recognition with CNNs Keras / Deep learning frameworks about / Keras design principles / Keras Keras implementation, of SqueezeNet reference / SqueezeNet Keras programming models sequential model / Keras programming models, Sequential model functional APIs / Keras programming models, Functional API Keras v2.1.4 reference / Functional API L latent factors (LFs) about / Model-based collaborative filtering LeNet5 about / LeNet5 implementing / Implementing a LeNet-5 step by step AlexNet / AlexNet transfer learning / Transfer learning pre-trained AlexNet / Pretrained AlexNet linear combination about / The artificial neuron linear regression about / Linear regression and beyond for real dataset / Linear regression revisited for a real dataset Long Short-Term Memory (LSTM) / Regularization Long Short-term memory units (LSTMs) about / Recurrent Neural Networks (RNNs) loss about / Fine-tuning implementation loss_val about / Emotion recognition with CNNs LSTM networks / LSTM networks LSTM predictive model, for sentiment analysis about / An LSTM predictive model for sentiment analysis network design / Network design LSTM model training / LSTM model training visualization, through TensorBoard / Visualizing through TensorBoard LSTM model evaluation / LSTM model evaluation M machine learning about / A soft introduction to machine learning supervised learning / Supervised learning unsupervised learning / Unsupervised learning reinforcement learning / Reinforcement learning Markov Chain Monte Carlo (MCMC) / Restricted Boltzmann Machines (RBMs) Markov Random Fields (MRF) about / Deep Belief Networks (DBNs) Mean Squared Error (MSE) / Autoencoder as an unsupervised feature learning algorithm mean square error (MSE) about / Linear regression and beyond Microsoft Cognitive Toolkit reference / Deep learning frameworks Microsoft Cognitive Toolkit (CNTK) / Deep learning frameworks model-based collaborative filtering about / Model-based collaborative filtering model inferencing about / A soft introduction to machine learning model parallelism about / Model parallelism MovieLens dataset about / Description of the dataset ratings data / Ratings data movies data / Movies data users data / Users data exploratory analysis / Exploratory analysis of the MovieLens dataset MovieLens website URL / Description of the dataset movie recommendation, with collaborative filtering developing / Movie recommendation using collaborative filtering utility matrix / The utility matrix dataset / Description of the dataset exploratory analysis, of MovieLens dataset / Exploratory analysis of the MovieLens dataset implementing / Implementing a movie RE evaluating / Evaluating the recommender system movie RE implementation performing / Implementing a movie RE model, training with available ratings / Training the model with the available ratings saved model, inferencing / Inferencing the saved model user-item table, generating / Generating the user-item table similar movies, clustering / Clustering similar movies movie rating prediction, by users / Movie rating prediction by users top k movies, finding / Finding top k movies top k similar movies, predicting / Predicting top k similar movies user-user similarity, computing / Computing user-user similarity Multi-Dimensional Recurrent Neural Networks (MD-RNNs) about / Emergent architectures multilayer perceptron about / Multilayer perceptron multilayer perceptron (MLP) implementing / Implementing a multilayer perceptron (MLP) training / Training an MLP using / Using MLPs dataset description / Dataset description preprocessing / Preprocessing TensorFlow implementation, for client-subscription assessment / A TensorFlow implementation of MLP for client-subscription assessment Multilayer Perceptron (MLP) / Deep Neural Networks (DNNs) MXNet / Deep learning frameworks N Natural Language Processing (NLP) / Introducing TensorFlow Lite Neon / Deep learning frameworks Neural Factorization Machines (NFMs) about / Neural factorization machines dataset description / Dataset description using, for movie recommendation / Using NFM for the movie recommendation FM model training / Model training NFM model training / Model training FM model, evaluating / Model evaluation NFM model, evaluating / Model evaluation neural network architectures about / Neural network architectures Deep Neural Networks (DNNs) / Deep Neural Networks (DNNs) Convolutional Neural Networks (CNNs) / Convolutional Neural Networks (CNNs) autoencoders / AutoEncoders Recurrent Neural Networks (RNNs) / Recurrent Neural Networks (RNNs) neuron about / The biological neurons normalization Z-score / Normalization min-max scaling / Normalization NVIDIA CUDA toolkit reference / Installing and configuring TensorFlow NVIDIA cuDNN reference / Installing and configuring TensorFlow NVIDIA GPU Cloud (NGC) / Deep learning frameworks NVIDIA Graph Analytics Library / Installing and configuring TensorFlow O OpenAI environments classic control and toy text / OpenAI environments algorithmic / OpenAI environments Atari / OpenAI environments board games / OpenAI environments 2D and 3D robots / OpenAI environments OpenAI Gym about / OpenAI Gym OpenAI environments / OpenAI environments env class / The env class installing / Installing and running OpenAI Gym URL / Installing and running OpenAI Gym running / Installing and running OpenAI Gym Open Neural Network Exchange (ONNX) / Deep learning frameworks output_pred about / Emotion recognition with CNNs overfitting about / A soft introduction to machine learning P parameter server about / Data parallelism PIL (Pillow) about / Pretrained AlexNet pixels about / Main concepts of CNNs pixel shaders about / The GPGPU history placeholders / TensorFlow code structure pre-trained AlexNet about / Pretrained AlexNet pre-trained VGG-19 neural network about / Content extractor and loss predictive model, for time series developing / Developing a predictive model for time series data dataset description / Description of the dataset pre-processing / Pre-processing and exploratory analysis exploratory analysis / Pre-processing and exploratory analysis LSTM predictive model / LSTM predictive model model evaluation / Model evaluation PrettyTensor about / PrettyTensor chaining layers / Chaining layers normal mode / Normal mode sequential mode / Sequential mode branch method / Branch and join join method / Branch and join digit classifier / Digit classifier Q Q-Learning algorithm about / The Q-Learning algorithm FrozenLake environment / The FrozenLake environment for FrozenLake problem / The FrozenLake environment Q-table / Deep Q-Network for the Cart-Pole problem R RandomizedSearchCV reference / GridSearch and randomized search for hyperparameters tuning read_data function about / Emotion recognition with CNNs receptive field / CNNs in action recommendation systems about / Recommendation systems collaborative filtering approaches, using / Collaborative filtering approaches Recurrent Neural Networks (RNNs) about / Neural network architectures, Recurrent Neural Networks (RNNs) working principles / Working principles of RNNs long-term dependency problem / RNN and the long-term dependency problem bi-directional RNNs / Bi-directional RNNs Bi-directional RNNs / Bi-directional RNNs gradient vanishing-exploding problem / RNN and the gradient vanishingexploding problem LSTM networks / LSTM networks Gated Recurrent Unit (GRU) cell / GRU cell Recurrent Neural Networks (RNNs), for spam prediction implementing / Implementing an RNN for spam prediction data description / Data description and preprocessing data preprocessing / Data description and preprocessing regression / Supervised learning regularization L2 regularization / Regularization L1 regularization / Regularization max-norm constraints / Regularization reinforcement learning about / Reinforcement learning ReLU using / Using ReLU ReLU operator / Implementing a LeNet-5 step by step residuals reference / Autoencoder as an unsupervised feature learning algorithm Restricted Boltzmann Machines (RBMs) about / Deep Belief Networks (DBNs) / Restricted Boltzmann Machines (RBMs) RL problem about / The RL problem RMSPropOptimizer about / Implementing a LeNet-5 step by step S sequential model, Keras sentiment classification, of movie reviews / Sentiment classification of movie reviews shared memory about / The CUDA architecture sigmoid about / The artificial neuron using / Using sigmoid Singular Value Decomposition (SVD) about / Hybrid recommender systems softmax activation function / CNNs in action softmax classifier / Softmax classifier softmax function / Activation functions using / Using softmax SqueezeNet / SqueezeNet Stacked Auto-Encoder (SAE) / Deep Neural Networks (DNNs) Staked Auto-Encoders (SAEs) about / Feed-forward neural networks (FFNNs) step function about / The artificial neuron Stochastic Gradient Descent (SDG) about / Stochastic gradient descent Stochastic Gradient Descent (SGD) / Autoencoder as an unsupervised feature learning algorithm streaming multiprocessor (SM) about / The CUDA architecture summary_op about / Emotion recognition with CNNs supervised learning about / Supervised learning synapses about / The biological neurons synaptic terminals about / The biological neurons T using / Using telodendria about / The biological neurons TensorBoard computations, visualizing / Visualizing computations through TensorBoard working / How does TensorBoard work? TensorFlow / Deep learning frameworks overview / A general overview of TensorFlow features, by latest release / A general overview of TensorFlow reference / A general overview of TensorFlow installing / Installing and configuring TensorFlow configuring / Installing and configuring TensorFlow computational graph / TensorFlow computational graph edges / TensorFlow computational graph code structure / TensorFlow code structure data model / Data model in TensorFlow about / Fine-tuning implementation Inception, exploring with / Exploring Inception with TensorFlow autoencoder, implementing / Implementing autoencoders with TensorFlow basic RNNs, implementing in / Implementing basic RNNs in TensorFlow TensorFlow GPU setup about / The TensorFlow GPU setup TensorFlow, updating / Update TensorFlow GPU representation / GPU representation GPU, using / Using a GPU GPU memory management / GPU memory management single GPU, assigning on multi-GPU system / Assigning a single GPU on a multi-GPU system source code, for GPU / The source code for GPU with soft placement multiple GPUs, using / Using multiple GPUs TensorFlow graph tf.Operation objects / TensorFlow computational graph tf.Tensor objects / TensorFlow computational graph components / TensorFlow computational graph TensorFlow Lite about / Introducing TensorFlow Lite TensorFlow v1.6 about / What's new from TensorFlow v1.6 forwards? Nvidia GPU support optimized / Nvidia GPU support optimized eager execution / Eager execution optimized accelerated linear algebra (XLA) / Optimized Accelerated Linear Algebra (XLA) tensors reference / TensorFlow code structure, Tensor about / Tensor test set about / A soft introduction to machine learning tf.estimator about / tf.estimator graph actions / Graph actions resources, parsing / Parsing resources flower predictions / Flower predictions TFLearn about / TFLearn layers / TFLearn graph_actions / TFLearn estimator / TFLearn installing / Installation Titanic survival predictor / Titanic survival predictor Theano / Deep learning frameworks TITO (tensor-in-tensor-out) / TensorFlow computational graph Torch / Deep learning frameworks training set about / A soft introduction to machine learning train_op about / Emotion recognition with CNNs transfer learning about / Transfer learning U unbalanced data about / Unbalanced data unsupervised learning about / Unsupervised learning utility matrix about / The utility matrix V validation set about / A soft introduction to machine learning vector space model reference / Preprocessing VGG about / VGG VGG-n about / VGG W weight optimization about / Weight optimization workers about / Data parallelism X Xavier initialization / Weights and biases ... Started with Deep Learning A soft introduction to machine learning Supervised learning Unbalanced data Unsupervised learning Reinforcement learning What is deep learning? Artificial neural networks. .. sector, and the main suspect behind these important developments is called deep learning Deep learning comprises a vast set of algorithms that are based on the concept of neural networks and expand... the way to the use of deep learning In this context, we propose the second edition of this book, with expanded and revised contents that introduce the core concepts of deep learning, using the