Book Image

Deep Learning By Example

Book Image

Deep Learning By Example

Overview of this book

Deep learning is a popular subset of machine learning, and it allows you to build complex models that are faster and give more accurate predictions. This book is your companion to take your first steps into the world of deep learning, with hands-on examples to boost your understanding of the topic. This book starts with a quick overview of the essential concepts of data science and machine learning which are required to get started with deep learning. It introduces you to Tensorflow, the most widely used machine learning library for training deep learning models. You will then work on your first deep learning problem by training a deep feed-forward neural network for digit classification, and move on to tackle other real-world problems in computer vision, language processing, sentiment analysis, and more. Advanced deep learning models such as generative adversarial networks and their applications are also covered in this book. By the end of this book, you will have a solid understanding of all the essential concepts in deep learning. With the help of the examples and code provided in this book, you will be equipped to train your own deep learning models with more confidence.

Preface

Who this book is for

What this book covers

To get the most out of this book

Free Chapter

Data Science - A Birds' Eye View

Data Science - A Birds' Eye View

Understanding data science by an example

Design procedure of data science algorithms

Getting to learn

Implementing the fish recognition/detection model

Different learning types

Data size and industry needs

Data Modeling in Action - The Titanic Example

Data Modeling in Action - The Titanic Example

Linear models for regression

Linear models for classification

Titanic example – model building and training

Different types of errors

Apparent (training set) error

Generalization/true error

Feature Engineering and Model Complexity – The Titanic Example Revisited

Feature Engineering and Model Complexity – The Titanic Example Revisited

Feature engineering

The curse of dimensionality

Titanic example revisited – all together

Bias-variance decomposition

Learning visibility

Get Up and Running with TensorFlow

Get Up and Running with TensorFlow

TensorFlow installation

The TensorFlow environment

Computational graphs

TensorFlow data types, variables, and placeholders

Getting output from TensorFlow

TensorBoard – visualizing learning

TensorFlow in Action - Some Basic Examples

TensorFlow in Action - Some Basic Examples

Capacity of a single neuron

Activation functions

Feed-forward neural network

The need for multilayer networks

TensorFlow terminologies – recap

Linear regression model – building and training

Logistic regression model – building and training

Deep Feed-forward Neural Networks - Implementing Digit Classification

Deep Feed-forward Neural Networks - Implementing Digit Classification

Hidden units and architecture design

MNIST dataset analysis

Digit classification – model building and training

Introduction to Convolutional Neural Networks

Introduction to Convolutional Neural Networks

The convolution operation

Different layers of CNNs

CNN basic example – MNIST digit classification

Object Detection – CIFAR-10 Example

Object Detection – CIFAR-10 Example

Object detection

CIFAR-10 – modeling, building, and training

Object Detection – Transfer Learning with CNNs

Object Detection – Transfer Learning with CNNs

Transfer learning

CIFAR-10 object detection – revisited

Recurrent-Type Neural Networks - Language Modeling

Recurrent-Type Neural Networks - Language Modeling

The intuition behind RNNs

Implementation of the language model

Representation Learning - Implementing Word Embeddings

Representation Learning - Implementing Word Embeddings

Introduction to representation learning

A practical example of the skip-gram architecture

Skip-gram Word2Vec implementation

Neural Sentiment Analysis

Neural Sentiment Analysis

General sentiment analysis architecture

Sentiment analysis – model implementation

Autoencoders – Feature Extraction and Denoising

Autoencoders – Feature Extraction and Denoising

Introduction to autoencoders

Examples of autoencoders

Autoencoder architectures

Compressing the MNIST dataset

Convolutional autoencoder

Denoising autoencoders

Applications of autoencoders

Generative Adversarial Networks

Generative Adversarial Networks

An intuitive introduction

Simple implementation of GANs

Face Generation and Handling Missing Labels

Face Generation and Handling Missing Labels

Face generation

Semi-supervised learning with Generative Adversarial Networks (GANs)

Implementing Fish Recognition

Implementing Fish Recognition

Code for fish recognition

Other Books You May Enjoy

Other Books You May Enjoy

Leave a review - let other readers know what you think

Customer Reviews

5 star

0

4 star

0

3 star

0

2 star

0

1 star

0

Autoencoder architectures

As we mentioned, a typical autoencoder consists of three parts. Let's explore these three parts in more detail. To motivate you, we are not going to reinvent the wheel here in this chapter. The encoder-decoder part is nothing but a fully connected neural network, and the code part is another neural network but it's not fully connected. The dimensionality of this code part is controllable and we can treat it as a hyperparameter:

Figure 3: General encoder-decoder architecture of autoencoders

Before diving into using autoencoders for compressing the MNIST dataset, we are going to list the set of hyperparameters that we can use to fine-tune the autoencoder model. There are mainly four hyperparameters:

Code part size: This is the number of units in the middle layer. The lower the number of units we have in this layer, the more compressed the representation...