book

Learning TensorFlow

by Tom Hope, Yehezkel S. Resheff, Itay Lieder

August 2017

Intermediate to advanced

240 pages

6h 2m

English

O'Reilly Media, Inc.

Read now

Unlock full access

Going DeepUsing TensorFlow for AI SystemsTensorFlow: What’s in a Name?A High-Level OverviewSummary
Installing TensorFlowHello WorldMNISTSoftmax RegressionSummary
Computation GraphsWhat Is a Computation Graph?The Benefits of Graph ComputationsGraphs, Sessions, and FetchesCreating a GraphCreating a Session and Running ItConstructing and Managing Our GraphFetchesFlowing TensorsNodes Are Operations, Edges Are Tensor ObjectsData TypesTensor Arrays and ShapesNamesVariables, Placeholders, and Simple OptimizationVariablesPlaceholdersOptimizationSummary
Introduction to CNNsMNIST: Take IIConvolutionPoolingDropoutThe ModelCIFAR10Loading the CIFAR10 DatasetSimple CIFAR10 ModelsSummary
The Importance of Sequence DataIntroduction to Recurrent Neural NetworksVanilla RNN ImplementationTensorFlow Built-in RNN FunctionsRNN for Text SequencesText SequencesSupervised Word EmbeddingsLSTM and Using Sequence LengthTraining Embeddings and the LSTM ClassifierSummary
Introduction to Word EmbeddingsWord2vecSkip-GramsEmbeddings in TensorFlowThe Noise-Contrastive Estimation (NCE) Loss FunctionLearning Rate DecayTraining and Visualizing with TensorBoardChecking Out Our EmbeddingsPretrained Embeddings, Advanced RNNPretrained Word EmbeddingsBidirectional RNN and GRU CellsSummary
Chapter OverviewHigh-Level Surveycontrib.learnLinear RegressionDNN ClassifierFeatureColumnHomemade CNN with contrib.learnTFLearnInstallationCNNRNNKerasPretrained models with TF-SlimSummary
The Input PipelineTFRecordsWriting with TFRecordWriterQueuesEnqueuing and DequeuingMultithreadingCoordinator and QueueRunnerA Full Multithreaded Input Pipelinetf.train.string_input_producer() and tf.TFRecordReader()tf.train.shuffle_batch()tf.train.start_queue_runners() and Wrapping UpSummary
Distributed ComputingWhere Does the Parallelization Take Place?What Is the Goal of Parallelization?TensorFlow Elementstf.app.flagsClusters and ServersReplicating a Computational Graph Across DevicesManaged SessionsDevice PlacementDistributed ExampleSummary

Saving and Exporting Our ModelAssigning Loaded WeightsThe Saver ClassIntroduction to TensorFlow ServingOverviewInstallationBuilding and ExportingSummary
Model Structuring and CustomizationModel StructuringCustomizationRequired and Recommended Components for TensorFlow ServingWhat Is a Docker Container and Why Do We Use It?Some Basic Docker Commands

Overview

Roughly inspired by the human brain, deep neural networks trained with large amounts of data can solve complex tasks with unprecedented accuracy. This practical book provides an end-to-end guide to TensorFlow, the leading open source software library that helps you build and train neural networks for computer vision, natural language processing (NLP), speech recognition, and general predictive analytics.

Authors Tom Hope, Yehezkel Resheff, and Itay Lieder provide a hands-on approach to TensorFlow fundamentals for a broad technical audience—from data scientists and engineers to students and researchers. You’ll begin by working through some basic examples in TensorFlow before diving deeper into topics such as neural network architectures, TensorBoard visualization, TensorFlow abstraction libraries, and multithreaded input pipelines. Once you finish this book, you’ll know how to build and deploy production-ready deep learning systems in TensorFlow.