Book Overview & Buying
Table Of Contents
Feedback & Rating

Neural Network Programming with TensorFlow

By : Dua, Ghotra

1.5 (2)

Buy this Book

Neural Network Programming with TensorFlow

1.5 (2)

By: Dua, Ghotra

Buy this Book

Overview of this book

If you're aware of the buzz surrounding the terms such as "machine learning," "artificial intelligence," or "deep learning," you might know what neural networks are. Ever wondered how they help in solving complex computational problem efficiently, or how to train efficient neural networks? This book will teach you just that. You will start by getting a quick overview of the popular TensorFlow library and how it is used to train different neural networks. You will get a thorough understanding of the fundamentals and basic math for neural networks and why TensorFlow is a popular choice Then, you will proceed to implement a simple feed forward neural network. Next you will master optimization techniques and algorithms for neural networks using TensorFlow. Further, you will learn to implement some more complex types of neural networks such as convolutional neural networks, recurrent neural networks, and Deep Belief Networks. In the course of the book, you will be working on real-world datasets to get a hands-on understanding of neural network programming. You will also get to train generative models and will learn the applications of autoencoders. By the end of this book, you will have a fair understanding of how you can leverage the power of TensorFlow to train neural networks of varying complexities, without any hassle. While you are learning about various neural network implementations you will learn the underlying mathematics and linear algebra and how they map to the appropriate TensorFlow constructs.

Preface

What this book covers

What you need for this book

Who this book is for

Conventions

Reader feedback

Customer support

Free Chapter

Maths for Neural Networks

Understanding linear algebra

Calculus

Optimization

Summary

Deep Feedforward Networks

Defining feedforward networks

Understanding backpropagation

Implementing feedforward networks with TensorFlow

Analyzing the Iris dataset

Implementing feedforward networks with images

Summary

Optimization for Neural Networks

What is optimization?

Types of optimizers

Gradient descent

Which optimizer to choose

Summary

Convolutional Neural Networks

An overview and the intuition of CNN

Convolution operations

Pooling

Image classification with convolutional networks

Summary

Recurrent Neural Networks

Introduction to RNNs

Introduction to long short term memory networks

Sentiment analysis

Summary

Generative Models

Generative models

GANs

Summary

Deep Belief Networking

Understanding deep belief networks

Model training

Predicting the label

Finding the accuracy of the model

DBN implementation for the MNIST dataset

Effect of the number of neurons in an RBM layer in a DBN

DBNs with two RBM layers

Classifying the NotMNIST dataset with a DBN

Summary

Autoencoders

Autoencoder algorithms

Under-complete autoencoders

Dataset

Basic autoencoders

Additive Gaussian Noise autoencoder

Sparse autoencoder

Summary

Research in Neural Networks

Avoiding overfitting in neural networks

Large-scale video processing with neural networks

Named entity recognition using a twisted neural network

Bidirectional RNNs

Summary

Getting started with TensorFlow

Environment setup

TensorFlow comparison with Numpy

Computational graph

Auto differentiation

TensorBoard

Customer Reviews

1.5 (2)

5 star

4 star

3 star

2 star

50%

1 star

50%

Types of optimizers

First, we look at the high-level categories of optimization algorithms and then dive deep into the individual optimizers.

First order optimization algorithms minimize or maximize a loss function using its gradient values concerning the parameters. The popularly used First order optimization algorithm is gradient descent. Here, the first order derivative tells us whether the function is decreasing or increasing at a particular point. The first order derivative gives us a line which is tangential to a point on its error surface.

The derivative for a function depends on single variables, whereas a gradient for a function depends on multiple variables.

Second order optimization algorithms use the second order derivative, which is also known as Hessian, to minimize or maximize the given loss function. Here, the Hessian is a matrix of second order partial derivatives...