book

Deep Learning for Beginners

Name: Deep Learning for Beginners
Author: Dr. Pablo Rivas
ISBN: 9781838640859

by Dr. Pablo Rivas

September 2020

Intermediate to advanced

432 pages

10h 41m

English

Packt Publishing

Read now

Unlock full access

Title Page
Copyright and Credits
Deep Learning for Beginners
About Packt
Why subscribe?
Foreword
Contributors
About the authorAbout the reviewersPackt is searching for authors like you
Preface
Who this book is forWhat this book coversTo get the most out of this bookDownload the example code filesDownload the color imagesConventions usedGet in touchReviews
Section 1: Getting Up to Speed
Introduction to Machine Learning
Diving into the ML ecosystemTraining ML algorithms from dataIntroducing deep learningThe model of a neuronThe perceptron learning algorithmShallow networksThe input-to-hidden layerThe hidden-to-hidden layerThe hidden-to-output layerDeep networksWhy is deep learning important today?Summary Questions and answersReferences
Setup and Introduction to Deep Learning Frameworks
Introduction to ColaboratoryIntroduction and setup of TensorFlowSetupTensorFlow with GPU supportPrinciples behind TensorFlowIntroduction and setup of KerasSetupPrinciples behind KerasIntroduction to PyTorchIntroduction to DopamineOther deep learning librariesCaffeTheanoHonorable mentionsSummary Questions and answersReferences
Preparing Data
Binary data and binary classificationBinary targets on the Cleveland Heart Disease datasetBinarizing the MNIST datasetBinarizing the imagesBinarizing the targetsCategorical data and multiple classesConverting string labels to numbersConverting categories to one-hot encodingReal-valued data and univariate regressionScaling to a specific range of valuesStandardizing to zero mean and unit varianceAltering the distribution of dataData augmentationRescalingAdding noiseRotatingOther augmentation techniquesData dimensionality reductionSupervised algorithmsLinear discriminant analysisUnsupervised techniquesKernel PCALarge datasetsSparse PCADictionary LearningRegarding the number of dimensionsEthical implications of manipulating dataSummary Questions and answersReferences

Learning from Data
Learning for a purposeClassificationBinary classificationMulti-class classificationRegressionMeasuring success and errorBinary classificationMultiple classesRegressionIdentifying overfitting and generalizationIf we have test dataNo test data? No problem – cross-validateThe art behind learningEthical implications of training deep learning algorithmsReporting using the appropriate performance measuresBeing careful with outliers and verifying themWeight classes with undersampled groupsSummary Questions and answersReferences
Training a Single Neuron
The perceptron modelThe visual conceptTensor operationsThe perceptron learning algorithmPLA in PythonA perceptron over non-linearly separable dataConvergence on linearly separable dataConvergence on non-linearly separable dataSummary Questions and answersReferences
Training Multiple Layers of Neurons
The MLP modelMinimizing the errorStep 1 – InitializationStep 2 – The forward passStep 3 – Calculating lossStep 4 – The backward passFinding the best hyperparametersSummary Questions and answersReferences
Section 2: Unsupervised Deep Learning
Autoencoders
Introduction to unsupervised learningEncoding and decoding layersEncoding layersDecoding layersLoss functionLearning and testingApplications in dimensionality reduction and visualizationMNIST data preparationAutoencoders for MNIST Training and visualizationEthical implications of unsupervised learningSummary Questions and answersReferences
Deep Autoencoders
Introducing deep belief networksMaking deep autoencodersBatch normalizationDropoutExploring latent spaces with deep autoencodersCIFAR-10MNIST Summary Questions and answersReferences
Variational Autoencoders
Introducing deep generative modelsExamining the VAE modelThe heart disease dataset revisitedThe re-parametrization trick and sampling Learning the posterior's parameters in the encoderModeling the decoderMinimizing the lossTraining a VAEGenerating data from the VAEComparing a deep and shallow VAE on MNISTShallow VAEDeep VAEEncoderDecoderDenoising VAEsThinking about the ethical implications of generative modelsSummary Questions and answersReferences
Restricted Boltzmann Machines
Introduction to RBMsBMsRBMsBernoulli RBMsLearning data representations with RBMsComparing RBMs and AEsSummaryQuestions and answersReferences
Section 3: Supervised Deep Learning
Deep and Wide Neural Networks
Wide neural networksDeep learning revisitedWide layersSummariesNamesThe CIFAR-10 datasetNew training toolsSaving or loading modelsReducing the learning rate on the flyStopping the learning process earlyResultsDense deep neural networksBuilding and training the modelResultsSparse deep neural networksBuilding a sparse network and training itResultsHyperparameter optimizationLibraries and parametersImplementation and resultsSummary Questions and answersReferences
Convolutional Neural Networks
Introduction to convolutional neural networksConvolution in n-dimensions1-dimension2-dimensionsn-dimensionsConvolutional layersConv2DThe layer+activation comboPooling strategiesConvolutional neural network for CIFAR-10ImplementationLoading dataCompiling the modelTraining the CNNResultsVisualization of filtersSummaryQuestions and answersReferences
Recurrent Neural Networks
Introduction to recurrent neural networksSimple RNNsEmbedding layersWord embedding and RNNs on IMDb Long short-term memory modelsSequence-to-vector modelsUnsupervised modelResultsVector-to-sequence modelsBi-directional LSTMImplementation and resultsSequence-to-sequence modelsEthical implicationsSummaryQuestions and answersReferences
Generative Adversarial Networks
Introducing adversarial learningLearning using an adversaryGANsTraining a GANAn MLP modelA convolutional modelComparing GANs and VAEsThinking about the ethical implications of GANsSummaryQuestions and answersReferences
Final Remarks on the Future of Deep Learning
Looking for advanced topics in deep learningDeep reinforcement learningSelf-supervised learningSystem 2 algorithmsLearning with more resources from PacktReinforcement learningSelf-supervised learningSummaryReferences
Other Books You May Enjoy
Leave a review - let other readers know what you think

Content preview from Deep Learning for Beginners

Deep Autoencoders

This chapter introduces the concept of deep belief networks and the significance of this type of deep unsupervised learning. It explains such concepts by introducing deep autoencoders along with two regularization techniques that can help create robust models. These regularization techniques, batch normalization and dropout, have been known to facilitate the learning of deep models and have been widely adopted. We will demonstrate the power of a deep autoencoder on MNIST and on a much harder dataset known as CIFAR-10, which contains color images.

By the end of this chapter, you will appreciate the benefits of making deep belief networks by observing the ease of modeling and quality of the output that they provide. You will ...

Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.

Read now

Unlock full access

More than 5,000 organizations count on O’Reilly

O’Reilly covers everything we've got, with content to help us build a world-class technology community, upgrade the capabilities and competencies of our teams, and improve overall team performance as well as their engagement.

Julian F.

Head of Cybersecurity

I wanted to learn C and C++, but it didn't click for me until I picked up an O'Reilly book. When I went on the O’Reilly platform, I was astonished to find all the books there, plus live events and sandboxes so you could play around with the technology.

Addison B.

Field Engineer

I’ve been on the O’Reilly platform for more than eight years. I use a couple of learning platforms, but I'm on O'Reilly more than anybody else. When you're there, you start learning. I'm never disappointed.

Amir M.

Data Platform Tech Lead

I'm always learning. So when I got on to O'Reilly, I was like a kid in a candy store. There are playlists. There are answers. There's on-demand training. It's worth its weight in gold, in terms of what it allows me to do.

Mark W.

Embedded Software Engineer

Publisher Resources

ISBN: 9781838640859

Cloud Computing

Data Engineering

Data Science

AI & ML

Programming Languages

Software Architecture

IT/Ops

Security

Design

Business

Soft Skills

Deep Learning for Beginners

by Dr. Pablo Rivas

Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.