Packt+ | Advance your knowledge in tech

You're reading from Deep Learning Quick Reference Useful hacks for training and optimizing deep neural networks with TensorFlow and Keras

Product type Paperback

Published in Mar 2018

Publisher Packt

ISBN-13 9781788837996

Length 272 pages

Edition 1st Edition

Languages

Python

Tools

Keras

Concepts

Deep Learning

Author (1):

Mike Bernico

View More author details

Table of Contents (22) Chapters

Title Page

Dedication

Packt Upsell

Foreword

Contributors

Preface

1. The Building Blocks of Deep Learning FREE CHAPTER

2. Using Deep Learning to Solve Regression Problems

3. Monitoring Network Training Using TensorBoard

4. Using Deep Learning to Solve Binary Classification Problems

5. Using Keras to Solve Multiclass Classification Problems

6. Hyperparameter Optimization

7. Training a CNN from Scratch

8. Transfer Learning with Pretrained CNNs

9. Training an RNN from scratch

10. Training LSTMs with Word Embeddings from Scratch

11. Training Seq2Seq Models

12. Using Deep Reinforcement Learning

13. Generative Adversarial Networks

1. Other Books You May Enjoy

Leave a review - let other readers know what you think

Index

Case study - handwritten digit classification

We will be using a multiclass classification network to recognize the corresponding class of a handwritten digit. As before, you can find the complete code for this chapter in the book's Git repository, under Chapter05, if you'd like to follow along.

Problem definition

The MNIST dataset has become an almost canonical neural network dataset. This dataset consists of images of 60,000 handwritten digits, belonging to 10 classes representing their respective digit (0,1,2...9). Because this dataset has become so common, many deep learning frameworks come with an MNIST loading method built into the API. Both TensorFlow and Keras have one, and we will be using the Keras MNIST loader to make our lives a little easier. However, should you want to obtain the data from its original source, or perhaps learn more about MNIST's history, you can find more at http://yann.lecun.com/exdb/mnist/.