Book Overview & Buying
Table Of Contents

Hands-On Convolutional Neural Networks with TensorFlow

By : Araujo, Zafar, Tzanidou, Burton, Patel

4 (3)

Buy this Book

Hands-On Convolutional Neural Networks with TensorFlow

4 (3)

By: Araujo, Zafar, Tzanidou, Burton, Patel

Buy this Book

Overview of this book

Convolutional Neural Networks (CNN) are one of the most popular architectures used in computer vision apps. This book is an introduction to CNNs through solving real-world problems in deep learning while teaching you their implementation in popular Python library - TensorFlow. By the end of the book, you will be training CNNs in no time! We start with an overview of popular machine learning and deep learning models, and then get you set up with a TensorFlow development environment. This environment is the basis for implementing and training deep learning models in later chapters. Then, you will use Convolutional Neural Networks to work on problems such as image classification, object detection, and semantic segmentation. After that, you will use transfer learning to see how these models can solve other deep learning problems. You will also get a taste of implementing generative models such as autoencoders and generative adversarial networks. Later on, you will see useful tips on machine learning best practices and troubleshooting. Finally, you will learn how to apply your models on large datasets of millions of images.

Preface

Who this book is for

What this book covers

To get the most out of this book

Get in touch

Free Chapter

Setup and Introduction to TensorFlow

The TensorFlow way of thinking

Setting up and installing TensorFlow

TensorFlow API levels

Building your first TensorFlow model

Creating TensorFlow graphs

Feeding data with placeholders

Initializing variables

Training our model

Evaluating a trained model

The session

Summary

Deep Learning and Convolutional Neural Networks

AI and ML

Artificial neural networks

Convolutional neural networks

Building a CNN model in TensorFlow

Summary

Image Classification in TensorFlow

CNN model architecture

Datasets

Image classification with TensorFlow

Model Initialization

Improving generalization by regularizing

Summary

Object Detection and Segmentation

Image classification with localization

Object detection as classification – Sliding window

Single Shot Detectors – You Only Look Once

Semantic segmentation

Instance segmentation

Summary

VGG, Inception Modules, Residuals, and MobileNets

Substituting big convolutions

VGGNet

GoogLeNet

Residual Networks

MobileNets

Summary

Autoencoders, Variational Autoencoders, and Generative Adversarial Networks

Why generative models

Autoencoders

Variational autoencoders

Generative adversarial networks

Summary

Transfer Learning

When?

How? An overview

How? Code example

Summary

Machine Learning Best Practices and Troubleshooting

Building Machine Learning Systems

Data Preparation

Bias and Variance

Data Imbalance

Evaluation Metrics

Code Structure best Practice

Recipe for CNN creation

Summary

Training at Scale

Storing data in TFRecords

Making efficient pipelines

Distributed computing in TensorFlow

When data does not fit on one computer

Scaling computation in the cloud

Summary

References

Chapter 1

Chapter 2

Chapter 3

Chapter 4

Chapter 5

Chapter 7

Chapter 9

Other Books You May Enjoy

Leave a review - let other readers know what you think

Why generative models

In the following illustration, we can see the main difference between generative models and discriminative models. With discriminative models, we generally try to find ways of separating or "discriminating" between different classes in our data. However, with generative models, we try to find out the probability distribution of our data. In the illustration, the distributions are represented by the large blue and yellow blobs that contain smaller circles. If we learn this distribution from our data, we will be able to sample or "generate" new data points that should belong to it like the red triangle.