Deep Learning with Python 1st Edition François

Chollet download

https://textbookfull.com/product/deep-learning-with-python-1st-
edition-francois-chollet/

Download more ebook from https://textbookfull.com

We believe these products will be a great fit for you. Click
the link to download now, or visit textbookfull.com
to discover even more!

Deep Learning with Python MEAP 2nd Edition Francois

Chollet

https://textbookfull.com/product/deep-learning-with-python-
meap-2nd-edition-francois-chollet/

Deep Learning with R 1st Edition François Chollet

https://textbookfull.com/product/deep-learning-with-r-1st-
edition-francois-chollet/

Probabilistic Deep Learning with Python 1st Edition

Oliver Duerr

https://textbookfull.com/product/probabilistic-deep-learning-
with-python-1st-edition-oliver-duerr/

Deep Learning with Python Learn Best Practices of Deep

Learning Models with PyTorch 2nd Edition Nikhil Ketkar
Jojo Moolayil

https://textbookfull.com/product/deep-learning-with-python-learn-
best-practices-of-deep-learning-models-with-pytorch-2nd-edition-
nikhil-ketkar-jojo-moolayil/
Deep Learning with Python Learn Best Practices of Deep
Learning Models with PyTorch 2nd Edition Nikhil Ketkar
Jojo Moolayil

https://textbookfull.com/product/deep-learning-with-python-learn-
best-practices-of-deep-learning-models-with-pytorch-2nd-edition-
nikhil-ketkar-jojo-moolayil-2/

Deep Learning with Python Develop Deep Learning Models

on Theano and TensorFLow Using Keras Jason Brownlee

https://textbookfull.com/product/deep-learning-with-python-
develop-deep-learning-models-on-theano-and-tensorflow-using-
keras-jason-brownlee/

Advanced Data Analytics Using Python: With Machine

Learning, Deep Learning and NLP Examples Mukhopadhyay

https://textbookfull.com/product/advanced-data-analytics-using-
python-with-machine-learning-deep-learning-and-nlp-examples-
mukhopadhyay/

Applied Neural Networks with TensorFlow 2: API Oriented

Deep Learning with Python Orhan Gazi Yalç■n

https://textbookfull.com/product/applied-neural-networks-with-
tensorflow-2-api-oriented-deep-learning-with-python-orhan-gazi-
yalcin/

Computer Vision Using Deep Learning Neural Network

Architectures with Python and Keras 1st Edition Vaibhav
Verdhan

https://textbookfull.com/product/computer-vision-using-deep-
learning-neural-network-architectures-with-python-and-keras-1st-
edition-vaibhav-verdhan/
Deep Learning with Python
François Chollet
Copyright

For online information and ordering of this and other Manning books,
please visit www.manning.com. The publisher offers discounts on this
book when ordered in quantity. For more information, please contact
Special Sales Department
Manning Publications Co.
20 Baldwin Road
PO Box 761
Shelter Island, NY 11964
Email: orders@manning.com

©2018 by Manning Publications Co. All rights reserved.

No part of this publication may be reproduced, stored in a retrieval

system, or transmitted, in any form or by means electronic,
mechanical, photocopying, or otherwise, without prior written
permission of the publisher.

Many of the designations used by manufacturers and sellers to

distinguish their products are claimed as trademarks. Where those
designations appear in the book, and Manning Publications was
aware of a trademark claim, the designations have been printed in
initial caps or all caps.

Recognizing the importance of preserving what has been written,

it is Manning’s policy to have the books we publish printed on acid-
free paper, and we exert our best efforts to that end. Recognizing
also our responsibility to conserve the resources of our planet,
Manning books are printed on paper that is at least 15 percent
recycled and processed without the use of elemental chlorine.
 Manning Publications Co.
20 Baldwin Road
PO Box 761
Shelter Island, NY 11964

Development editor: Toni Arritola

Technical development editor: Jerry Gaines
Review editor: Aleksandar Dragosavljević
Project editor: Tiffany Taylor
Copyeditor: Tiffany Taylor
Proofreader: Katie Tennant
Technical proofreaders: Alex Ott and Richard Tobias
Typesetter: Dottie Marsico
Cover designer: Marija Tudor

ISBN 9781617294433

Printed in the United States of America

1 2 3 4 5 6 7 8 9 10 – EBM – 22 21 20 19 18 17
Brief Table of Contents
Copyright
Brief Table of Contents
Table of Contents
Preface
Acknowledgments
About this Book
About the Author
About the Cover

1. Fundamentals of deep learning

Chapter 1. What is deep learning?

Chapter 2. Before we begin: the mathematical building blocks of
neural networks
Chapter 3. Getting started with neural networks
Chapter 4. Fundamentals of machine learning

2. Deep learning in practice

Chapter 5. Deep learning for computer vision

Chapter 6. Deep learning for text and sequences
Chapter 7. Advanced deep-learning best practices
Chapter 8. Generative deep learning
Chapter 9. Conclusions

Appendix A. Installing Keras and its dependencies on Ubuntu

Appendix B. Running Jupyter notebooks on an EC2 GPU instance
Index

List of Figures

List of Tables

List of Listings
Table of Contents
Copyright
Brief Table of Contents
Table of Contents
Preface
Acknowledgments
About this Book
About the Author
About the Cover

1. Fundamentals of deep learning

Chapter 1. What is deep learning?

1.1. Artificial intelligence, machine learning, and deep
learning

1.1.1. Artificial intelligence

1.1.2. Machine learning
1.1.3. Learning representations from data
1.1.4. The “deep” in deep learning
1.1.5. Understanding how deep learning works, in three
figures
1.1.6. What deep learning has achieved so far
1.1.7. Don’t believe the short-term hype
1.1.8. The promise of AI

1.2. Before deep learning: a brief history of machine learning

1.2.1. Probabilistic modeling

 1.2.2. Early neural networks
1.2.3. Kernel methods
1.2.4. Decision trees, random forests, and gradient
boosting machines
1.2.5. Back to neural networks
1.2.6. What makes deep learning different
1.2.7. The modern machine-learning landscape

1.3. Why deep learning? Why now?

1.3.1. Hardware
1.3.2. Data
1.3.3. Algorithms
1.3.4. A new wave of investment
1.3.5. The democratization of deep learning
1.3.6. Will it last?

Chapter 2. Before we begin: the mathematical building blocks of

neural networks

2.1. A first look at a neural network

2.2. Data representations for neural networks
2.2.1. Scalars (0D tensors)
2.2.2. Vectors (1D tensors)
2.2.3. Matrices (2D tensors)
2.2.4. 3D tensors and higher-dimensional tensors
2.2.5. Key attributes
2.2.6. Manipulating tensors in Numpy
2.2.7. The notion of data batches
 2.2.8. Real-world examples of data tensors
2.2.9. Vector data
2.2.10. Timeseries data or sequence data
2.2.11. Image data
2.2.12. Video data

2.3. The gears of neural networks: tensor operations

2.3.1. Element-wise operations

2.3.2. Broadcasting
2.3.3. Tensor dot
2.3.4. Tensor reshaping
2.3.5. Geometric interpretation of tensor operations
2.3.6. A geometric interpretation of deep learning

2.4. The engine of neural networks: gradient-based

optimization

2.4.1. What’s a derivative?

2.4.2. Derivative of a tensor operation: the gradient
2.4.3. Stochastic gradient descent
2.4.4. Chaining derivatives: the Backpropagation
algorithm

2.5. Looking back at our first example

Chapter 3. Getting started with neural networks

3.1. Anatomy of a neural network

3.1.1. Layers: the building blocks of deep learning

3.1.2. Models: networks of layers
 3.1.3. Loss functions and optimizers: keys to configuring
the learning process

3.2. Introduction to Keras

3.2.1. Keras, TensorFlow, Theano, and CNTK

3.2.2. Developing with Keras: a quick overview

3.3. Setting up a deep-learning workstation

3.3.1. Jupyter notebooks: the preferred way to run deep-

learning experiments
3.3.2. Getting Keras running: two options
3.3.3. Running deep-learning jobs in the cloud: pros and
cons
3.3.4. What is the best GPU for deep learning?

3.4. Classifying movie reviews: a binary classification

example
3.4.1. The IMDB dataset
3.4.2. Preparing the data
3.4.3. Building your network
3.4.4. Validating your approach
3.4.5. Using a trained network to generate predictions on
new data
3.4.6. Further experiments
3.4.7. Wrapping up

3.5. Classifying newswires: a multiclass classification example

3.5.1. The Reuters dataset
3.5.2. Preparing the data
 3.5.3. Building your network
3.5.4. Validating your approach
3.5.5. Generating predictions on new data
3.5.6. A different way to handle the labels and the loss
3.5.7. The importance of having sufficiently large
intermediate layers
3.5.8. Further experiments
3.5.9. Wrapping up

3.6. Predicting house prices: a regression example

3.6.1. The Boston Housing Price dataset

3.6.2. Preparing the data
3.6.3. Building your network
3.6.4. Validating your approach using K-fold validation
3.6.5. Wrapping up

Chapter 4. Fundamentals of machine learning

4.1. Four branches of machine learning

4.1.1. Supervised learning
4.1.2. Unsupervised learning
4.1.3. Self-supervised learning
4.1.4. Reinforcement learning

4.2. Evaluating machine-learning models

4.2.1. Training, validation, and test sets
4.2.2. Things to keep in mind
 4.3. Data preprocessing, feature engineering, and feature
learning
4.3.1. Data preprocessing for neural networks
4.3.2. Feature engineering

4.4. Overfitting and underfitting

4.4.1. Reducing the network’s size
4.4.2. Adding weight regularization
4.4.3. Adding dropout

4.5. The universal workflow of machine learning

4.5.1. Defining the problem and assembling a dataset
4.5.2. Choosing a measure of success
4.5.3. Deciding on an evaluation protocol
4.5.4. Preparing your data
4.5.5. Developing a model that does better than a baseline
4.5.6. Scaling up: developing a model that overfits
4.5.7. Regularizing your model and tuning your
hyperparameters

2. Deep learning in practice

Chapter 5. Deep learning for computer vision
5.1. Introduction to convnets

5.1.1. The convolution operation

5.1.2. The max-pooling operation

5.2. Training a convnet from scratch on a small dataset

 5.2.1. The relevance of deep learning for small-data
problems
5.2.2. Downloading the data
5.2.3. Building your network
5.2.4. Data preprocessing
5.2.5. Using data augmentation

5.3. Using a pretrained convnet

5.3.1. Feature extraction
5.3.2. Fine-tuning
5.3.3. Wrapping up

5.4. Visualizing what convnets learn

5.4.1. Visualizing intermediate activations

5.4.2. Visualizing convnet filters
5.4.3. Visualizing heatmaps of class activation

Chapter 6. Deep learning for text and sequences

6.1. Working with text data

6.1.1. One-hot encoding of words and characters

6.1.2. Using word embeddings
6.1.3. Putting it all together: from raw text to word
embeddings
6.1.4. Wrapping up

6.2. Understanding recurrent neural networks

6.2.1. A recurrent layer in Keras

6.2.2. Understanding the LSTM and GRU layers
 6.2.3. A concrete LSTM example in Keras
6.2.4. Wrapping up

6.3. Advanced use of recurrent neural networks

6.3.1. A temperature-forecasting problem

6.3.2. Preparing the data
6.3.3. A common-sense, non-machine-learning baseline
6.3.4. A basic machine-learning approach
6.3.5. A first recurrent baseline
6.3.6. Using recurrent dropout to fight overfitting
6.3.7. Stacking recurrent layers
6.3.8. Using bidirectional RNNs
6.3.9. Going even further
6.3.10. Wrapping up

6.4. Sequence processing with convnets

6.4.1. Understanding 1D convolution for sequence data

6.4.2. 1D pooling for sequence data
6.4.3. Implementing a 1D convnet
6.4.4. Combining CNNs and RNNs to process long
sequences
6.4.5. Wrapping up

Chapter 7. Advanced deep-learning best practices

7.1. Going beyond the Sequential model: the Keras functional

API

7.1.1. Introduction to the functional API

 7.1.2. Multi-input models
7.1.3. Multi-output models
7.1.4. Directed acyclic graphs of layers
7.1.5. Layer weight sharing
7.1.6. Models as layers
7.1.7. Wrapping up

7.2. Inspecting and monitoring deep-learning models using

Keras callbacks and TensorBoard

7.2.1. Using callbacks to act on a model during training

7.2.2. Introduction to TensorBoard: the TensorFlow
visualization framework
7.2.3. Wrapping up

7.3. Getting the most out of your models

7.3.1. Advanced architecture patterns
7.3.2. Hyperparameter optimization
7.3.3. Model ensembling
7.3.4. Wrapping up

Chapter 8. Generative deep learning

8.1. Text generation with LSTM

8.1.1. A brief history of generative recurrent networks

8.1.2. How do you generate sequence data?
8.1.3. The importance of the sampling strategy
8.1.4. Implementing character-level LSTM text generation
8.1.5. Wrapping up
 8.2. DeepDream

8.2.1. Implementing DeepDream in Keras

8.2.2. Wrapping up

8.3. Neural style transfer

8.3.1. The content loss
8.3.2. The style loss
8.3.3. Neural style transfer in Keras
8.3.4. Wrapping up

8.4. Generating images with variational autoencoders

8.4.1. Sampling from latent spaces of images

8.4.2. Concept vectors for image editing
8.4.3. Variational autoencoders
8.4.4. Wrapping up

8.5. Introduction to generative adversarial networks

8.5.1. A schematic GAN implementation
8.5.2. A bag of tricks
8.5.3. The generator
8.5.4. The discriminator
8.5.5. The adversarial network
8.5.6. How to train your DCGAN
8.5.7. Wrapping up

Chapter 9. Conclusions

9.1. Key concepts in review

 9.1.1. Various approaches to AI
9.1.2. What makes deep learning special within the field
of machine learning
9.1.3. How to think about deep learning
9.1.4. Key enabling technologies
9.1.5. The universal machine-learning workflow
9.1.6. Key network architectures
9.1.7. The space of possibilities

9.2. The limitations of deep learning

9.2.1. The risk of anthropomorphizing machine-learning

models
9.2.2. Local generalization vs. extreme generalization
9.2.3. Wrapping up

9.3. The future of deep learning

9.3.1. Models as programs
9.3.2. Beyond backpropagation and differentiable layers
9.3.3. Automated machine learning
9.3.4. Lifelong learning and modular subroutine reuse
9.3.5. The long-term vision

9.4. Staying up to date in a fast-moving field

9.4.1. Practice on real-world problems using Kaggle

9.4.2. Read about the latest developments on arXiv
9.4.3. Explore the Keras ecosystem

9.5. Final words

Appendix A. Installing Keras and its dependencies on Ubuntu

A.1. Installing the Python scientific suite

A.2. Setting up GPU support
A.3. Installing Theano (optional)
A.4. Installing Keras

Appendix B. Running Jupyter notebooks on an EC2 GPU instance

B.1. What are Jupyter notebooks? Why run Jupyter notebooks on
AWS GPUs?
B.2. Why would you not want to use Jupyter on AWS for deep
learning?
B.3. Setting up an AWS GPU instance

B.3.1. Configuring Jupyter

B.4. Installing Keras

B.5. Setting up local port forwarding
B.6. Using Jupyter from your local browser

Index

List of Figures

List of Tables

List of Listings
Preface

If you’ve picked up this book, you’re probably aware of the

extraordinary progress that deep learning has represented for the
field of artificial intelligence in the recent past. In a mere five years,
we’ve gone from near-unusable image recognition and speech
transcription, to superhuman performance on these tasks.

The consequences of this sudden progress extend to almost every

industry. But in order to begin deploying deep-learning technology to
every problem that it could solve, we need to make it accessible to
as many people as possible, including non-experts—people who
aren’t researchers or graduate students. For deep learning to reach
its full potential, we need to radically democratize it.

When I released the first version of the Keras deep-learning

framework in March 2015, the democratization of AI wasn’t what I
had in mind. I had been doing research in machine learning for
several years, and had built Keras to help me with my own
experiments. But throughout 2015 and 2016, tens of thousands of
new people entered the field of deep learning; many of them picked
up Keras because it was—and still is—the easiest framework to get
started with. As I watched scores of newcomers use Keras in
unexpected, powerful ways, I came to care deeply about the
accessibility and democratization of AI. I realized that the further we
spread these technologies, the more useful and valuable they
become. Accessibility quickly became an explicit goal in the
development of Keras, and over a few short years, the Keras
developer community has made fantastic achievements on this front.
We’ve put deep learning into the hands of tens of thousands of
people, who in turn are using it to solve important problems we
didn’t even know existed until recently.
The book you’re holding is another step on the way to making deep
learning available to as many people as possible. Keras had always
needed a companion course to simultaneously cover fundamentals
of deep learning, Keras usage patterns, and deep-learning best
practices. This book is my best effort to produce such a course. I
wrote it with a focus on making the concepts behind deep learning,
and their implementation, as approachable as possible. Doing so
didn’t require me to dumb down anything—I strongly believe that
there are no difficult ideas in deep learning. I hope you’ll find this
book valuable and that it will enable you to begin building intelligent
applications and solve the problems that matter to you.
Acknowledgments

I’d like to thank the Keras community for making this book possible.
Keras has grown to have hundreds of open source contributors and
more than 200,000 users. Your contributions and feedback have
turned Keras into what it is today.

I’d also like to thank Google for backing the Keras project. It has
been fantastic to see Keras adopted as TensorFlow’s high-level API.
A smooth integration between Keras and TensorFlow greatly benefits
both TensorFlow users and Keras users and makes deep learning
accessible to most.

I want to thank the people at Manning who made this book possible:
publisher Marjan Bace and everyone on the editorial and production
teams, including Christina Taylor, Janet Vail, Tiffany Taylor, Katie
Tennant, Dottie Marsico, and many others who worked behind the
scenes.

Many thanks go to the technical peer reviewers led by Aleksandar

Dragosavljević—Diego Acuña Rozas, Geoff Barto, David Blumenthal-
Barby, Abel Brown, Clark Dorman, Clark Gaylord, Thomas Heiman,
Wilson Mar, Sumit Pal, Vladimir Pasman, Gustavo Patino, Peter
Rabinovitch, Alvin Raj, Claudio Rodriguez, Srdjan Santic, Richard
Tobias, Martin Verzilli, William E. Wheeler, and Daniel Williams—and
the forum contributors. Their contributions included catching
technical mistakes, errors in terminology, and typos, and making
topic suggestions. Each pass through the review process and each
piece of feedback implemented through the forum topics shaped and
molded the manuscript.

On the technical side, special thanks go to Jerry Gaines, who served

as the book’s technical editor; and Alex Ott and Richard Tobias, who
served as the book’s technical proofreaders. They’re the best
technical editors I could have hoped for.

Finally, I’d like to express my gratitude to my wife Maria for being

extremely supportive throughout the development of Keras and the
writing of this book.
About this Book

This book was written for anyone who wishes to explore deep
learning from scratch or broaden their understanding of deep
learning. Whether you’re a practicing machine-learning engineer, a
software developer, or a college student, you’ll find value in these
pages.

This book offers a practical, hands-on exploration of deep learning.

It avoids mathematical notation, preferring instead to explain
quantitative concepts via code snippets and to build practical
intuition about the core ideas of machine learning and deep learning.

You’ll learn from more than 30 code examples that include detailed
commentary, practical recommendations, and simple high-level
explanations of everything you need to know to start using deep
learning to solve concrete problems.

The code examples use the Python deep-learning framework Keras,

with TensorFlow as a backend engine. Keras, one of the most
popular and fastest-growing deep-learning frameworks, is widely
recommended as the best tool to get started with deep learning.

After reading this book, you’ll have a solid understand of what deep
learning is, when it’s applicable, and what its limitations are. You’ll
be familiar with the standard workflow for approaching and solving
machine-learning problems, and you’ll know how to address
commonly encountered issues. You’ll be able to use Keras to tackle
real-world problems ranging from computer vision to natural-
language processing: image classification, timeseries forecasting,
sentiment analysis, image and text generation, and more.

Who should read this book

This book is written for people with Python programming experience
who want to get started with machine learning and deep learning.
But this book can also be valuable to many different types of
readers:

If you’re a data scientist familiar with machine learning, this

book will provide you with a solid, practical introduction to deep
learning, the fastest-growing and most significant subfield of
machine learning.
If you’re a deep-learning expert looking to get started with the
Keras framework, you’ll find this book to be the best Keras crash
course available.
If you’re a graduate student studying deep learning in a formal
setting, you’ll find this book to be a practical complement to
your education, helping you build intuition around the behavior
of deep neural networks and familiarizing you with key best
practices.

Even technically minded people who don’t code regularly will find
this book useful as an introduction to both basic and advanced deep-
learning concepts.

In order to use Keras, you’ll need reasonable Python proficiency.

Additionally, familiarity with the Numpy library will be helpful,
although it isn’t required. You don’t need previous experience with
machine learning or deep learning: this book covers from scratch all
the necessary basics. You don’t need an advanced mathematics
background, either—high school–level mathematics should suffice in
order to follow along.

Roadmap

This book is structured in two parts. If you have no prior experience

with machine learning, I strongly recommend that you complete part
1 before approaching part 2. We’ll start with simple examples, and as
the book goes on, we’ll get increasingly close to state-of-the-art
techniques.

Part 1 is a high-level introduction to deep learning, providing context

and definitions, and explaining all the notions required to get started
with machine learning and neural networks:

Chapter 1 presents essential context and background knowledge

around AI, machine learning, and deep learning.
Chapter 2 introduces fundamental concepts necessary in order to
approach deep learning: tensors, tensor operations, gradient
descent, and backpropagation. This chapter also features the
book’s first example of a working neural network.
Chapter 3 includes everything you need to get started with
neural networks: an introduction to Keras, our deep-learning
framework of choice; a guide for setting up your workstation;
and three foundational code examples with detailed
explanations. By the end of this chapter, you’ll be able to train
simple neural networks to handle classification and regression
tasks, and you’ll have a solid idea of what’s happening in the
background as you train them.
Chapter 4 explores the canonical machine-learning workflow.
You’ll also learn about common pitfalls and their solutions.

Part 2 takes an in-depth dive into practical applications of deep

learning in computer vision and natural-language processing. Many
of the examples introduced in this part can be used as templates to
solve problems you’ll encounter in the real-world practice of deep
learning:

Chapter 5 examines a range of practical computer-vision

examples, with a focus on image classification.
Chapter 6 gives you practice with techniques for processing
sequence data, such as text and timeseries.
 Chapter 7 introduces advanced techniques for building state-of-
the-art deep-learning models.
Chapter 8 explains generative models: deep-learning models
capable of creating images and text, with sometimes
surprisingly artistic results.
Chapter 9 is dedicated to consolidating what you’ve learned
throughout the book, as well as opening perspectives on the
limitations of deep learning and exploring its probable future.

Software/hardware requirements

All of this book’s code examples use the Keras deep-learning

framework (https://keras.io), which is open source and free to
download. You’ll need access to a UNIX machine; it’s possible to use
Windows, too, but I don’t recommend it. Appendix A walks you
through the complete setup.

I also recommend that you have a recent NVIDIA GPU on your

machine, such as a TITAN X. This isn’t required, but it will make
your experience better by allowing you to run the code examples
several times faster. See section 3.3 for more information about
setting up a deep-learning workstation.

If you don’t have access to a local workstation with a recent NVIDIA

GPU, you can use a cloud environment, instead. In particular, you
can use Google Cloud instances (such as an n1-standard-8 instance
with an NVIDIA Tesla K80 add-on) or Amazon Web Services (AWS)
GPU instances (such as a p2.xlarge instance). Appendix B presents in
detail one possible cloud workflow that runs an AWS instance via
Jupyter notebooks, accessible in your browser.

Source code
All code examples in this book are available for download as Jupyter
notebooks from the book’s website, www.manning.com/books/deep-
learning-with-python, and on GitHub at https://github.com/fchollet/deep-
learning-with-python-notebooks.

Book forum

Purchase of Deep Learning with Python includes free access to a

private web forum run by Manning Publications where you can make
comments about the book, ask technical questions, and receive help
from the author and from other users. To access the forum, go to
https://forums.manning.com/forums/deep-learning-with-python. You can
also learn more about Manning’s forums and the rules of conduct at
https://forums.manning.com/forums/about.

Manning’s commitment to our readers is to provide a venue where a

meaningful dialogue between individual readers and between
readers and the author can take place. It isn’t a commitment to any
specific amount of participation on the part of the author, whose
contribution to the forum remains voluntary (and unpaid). We
suggest you try asking him some challenging questions lest his
interest stray! The forum and the archives of previous discussions
will be accessible from the publisher’s website as long as the book is
in print.
About the Author

François Chollet works on deep learning at Google in Mountain View,

CA. He is the creator of the Keras deep-learning library, as well as a
contributor to the TensorFlow machine-learning framework. He also
does deep-learning research, with a focus on computer vision and
the application of machine learning to formal reasoning. His papers
have been published at major conferences in the field, including the
Conference on Computer Vision and Pattern Recognition (CVPR), the
Conference and Workshop on Neural Information Processing
Systems (NIPS), the International Conference on Learning
Representations (ICLR), and others.
About the Cover

The figure on the cover of Deep Learning with Python is captioned

“Habit of a Persian Lady in 1568.” The illustration is taken from
Thomas Jefferys’ A Collection of the Dresses of Different Nations,
Ancient and Modern (four volumes), London, published between
1757 and 1772. The title page states that these are hand-colored
copperplate engravings, heightened with gum arabic.

Thomas Jefferys (1719–1771) was called “Geographer to King

George III.” He was an English cartographer who was the leading
map supplier of his day. He engraved and printed maps for
government and other official bodies and produced a wide range of
commercial maps and atlases, especially of North America. His work
as a map maker sparked an interest in local dress customs of the
lands he surveyed and mapped, which are brilliantly displayed in this
collection. Fascination with faraway lands and travel for pleasure
were relatively new phenomena in the late eighteenth century, and
collections such as this one were popular, introducing both the
tourist as well as the armchair traveler to the inhabitants of other
countries.

The diversity of the drawings in Jefferys’ volumes speaks vividly of

the uniqueness and individuality of the world’s nations some 200
years ago. Dress codes have changed since then, and the diversity
by region and country, so rich at the time, has faded away. It’s now
often hard to tell the inhabitants of one continent from another.
Perhaps, trying to view it optimistically, we’ve traded a cultural and
visual diversity for a more varied personal life—or a more varied and
interesting intellectual and technical life.

At a time when it’s difficult to tell one computer book from another,
Manning celebrates the inventiveness and initiative of the computer
Exploring the Variety of Random
Documents with Different Content
did not understand what he had felt, nor could he see that it might
have significance. What was the 'interpretation' of a storm, of an
earthquake, or of winter and summer? God, perhaps; perhaps just
'nature.' He did not know. Margaret had told him the story of the
opera in the evening; he had followed it easily enough and could not
forget it. It was a sort of religious fairy-tale, he thought, and he was
ready to believe that Wagner had made a good poem of it, even a
great poem. But it was not the story that could be told, which had
moved him; it was nothing so easily defined as a poem, or a drama,
or a piece of music. A far more cultivated man than he could ever
become might sit through the performance and feel little or nothing,
of that he was sure; just as he could have carried beautiful Lady
Maud in his arms without feeling that she was a woman for him,
whereas the slightest touch of Margaret Donne, the mere fact of
being near her, made the blood beat in his throat.
That was only a way of putting it, for there was no sex in the music
he had just heard. He had sat so close to Margaret that their arms
constantly touched, yet he had forgotten that she was there. If the
music had been Tristan and Isolde he could not have been unaware
of her, for a moment, for that is the supreme sex-music of Wagner's
art. But this was different, altogether different, though it was even
stronger than that.
He forgot to look at his watch. Margaret came out of the hotel,
expecting to find him waiting for her within the hall, and prepared to
be annoyed with him for taking so long over a meal. She stood on
the step and looked about, and saw him sitting on the bench at a
little distance. He raised his eyes as she came towards him and then
rose quickly.
'Is it time?' he asked.
'Yes,' she said. 'Did you get anything decent to eat?'
'Yes,' he answered vaguely. 'That is, now I think of it, I forgot about
dinner. It doesn't matter.'
She looked at his hard face curiously and saw a dead blank, the
blank that had sometimes frightened her by its possibilities, when
the eyes alone came suddenly to life.
'Won't you go in and get a biscuit, or a sandwich?' she asked after a
moment.
'Oh, no, thanks. I'm used to skipping meals when I'm interested in
things. Let's go, if you're ready.'
'I believe you are one of nature's Wagnerites,' Margaret said, as they
drove up the hill again, and she smiled at the idea.
'Well,' he answered slowly, 'there's one thing, if you don't mind my
telling you. It's rather personal. Perhaps I'd better not.'
The Primadonna was silent for a few moments, and did not look at
him.
'Tell me,' she said suddenly.
'It's this. I don't know how long the performance lasted, but while it
was going on I forgot you were close beside me. You might just as
well not have been there. It's the first time since I ever knew you
that I've been near you without thinking about you all the time, and
I hadn't realised it till I was sitting here by myself. I hope you don't
mind my telling you?'
'It only makes me more glad that I brought you,' Margaret said
quietly.
'Thank you,' he answered; but he was quite sure that the same thing
could not happen again during the Second Part.
Nevertheless, it happened. For a little while, they were man and
woman, sitting side by side and very near, two in a silent multitude
of other men and women; but before long he was quite motionless,
his eyes were fixed again and he had forgotten her. She saw it and
wondered, for she knew how her presence moved him, and as his
hands lay folded on his knee, a mischievous girlish impulse almost
made her, the great artist, forget that she was listening to the
greatest music in the world and nearly made her lay her hand on
his, just to see what he would do. She was ashamed of it, and a
little disgusted with herself. The part of her that was Margaret
Donne felt the disgust; the part that was Cordova felt the shame,
and each side of her nature was restrained at a critical moment. Yet
when the 'Good Friday' music began, she was thinking of Van Torp
and he was unconscious of her presence.
It could not last, and soon she, too, was taken up into the artificial
paradise of the master-musician and borne along in the gale of
golden wings, and there was no passing of time till the very end;
and the people rose in silence and went out under the summer
stars; and all those whom nature had gifted to hear rightly, took
with them memories that years would scarcely dim.
The two walked slowly back to the town as the crowd scattered on
foot and in carriages. It was warm, and there was no moon, and one
could smell the dust, for many people were moving in the same
direction, though some stopped at almost every house and went in,
and most of them were beginning to talk in quiet tones.
Margaret stepped aside from the road and entered a narrow lane,
and Van Torp followed her in silence.
'This leads out to the fields,' she said. 'I must breathe the fresh air.
Do you mind?'
'On the contrary.'
 "She was aware of his slight change of
position without turning her eyes."

He said nothing more, and she did not speak, but walked on without
haste, dilating her nostrils to the sweet smell of grass that reached
her already. In a little while they had left the houses behind them,
and they came to a gate that led into a field.
Van Torp was going to undo the fastening, for there was no lock.
'No,' she said, 'we won't go through. I love to lean on a gate.'
She rested her crossed arms on the upper rail and Van Torp did the
same, careful that his elbow should not touch hers, and they both
stared into the dim, sweet-scented meadow. He felt her presence
now and it almost hurt him; he could hear his slow pulse in his ears,
hard and regular. She did not speak, but the night was so still that
he could hear her breathing, and at last he could not bear the warm
silence any longer.
'What are you thinking about?' he asked, trying to speak lightly.
She waited, or hesitated, before she answered him.
'You,' she said, after a time.
He moved involuntarily, and then drew a little further away from her,
as he might have withdrawn a foot from the edge of a precipice, out
of common caution. She was aware of his slight change of position
without turning her eyes.
'What made you say what you did to Mrs. Rushmore yesterday
afternoon?' she asked.
'About you?'
'Yes.'
'She asked me, point-blank, what I thought of Logotheti,' Van Torp
answered. 'I told her that I couldn't give her an unbiassed opinion of
the man you meant to marry, because I had always hoped to marry
you myself.'
'Oh—was that the way it happened?'
'Mrs. Rushmore could hardly have misunderstood me,' said Van Torp,
gathering the reins of himself, so to say, for anything that might
happen.
'No. But it sounds differently when you say it yourself.'
'That was just what I said, anyhow,' answered Van Torp. 'I didn't
think she'd go and tell you right away, but since she has, I don't
regret having said that much.'
'It was straightforward, at all events—if it was all true!' There was
the faintest laugh in her tone as she spoke the last words.
'It's true, right enough, though I didn't expect that I should be
talking to you about this sort of thing to-night.'
'The effect on Mrs. Rushmore was extraordinary, positively
fulminating,' Margaret said more lightly. 'She says I ought to break
off my engagement at once, and marry you! Fancy!'
'That's very kind of her, I'm sure,' observed Mr. Van Torp.
'I don't think so. I like it less and less, the more I think of it.'
'Well, I'm sorry, but I suppose it's natural, since you've concluded to
marry him, and it can't be helped. I wasn't going to say anything
against him, and I wouldn't say anything for him, so there was
nothing to do but to explain, which I did. I'm sorry you think I did
wrong, but I should give the same answer again.'
'Mrs. Rushmore thinks that Konstantin is a designing foreigner
because he's a Greek man of business, and that you are perfection
because you are an American business man.'
'If I'm perfection, that's not the real reason,' said Van Torp,
snatching at his first chance to steer out of the serious current; but
Margaret did not laugh.
'You are not perfection, nor I either,' she answered gravely. 'You are
famous in your way, and people call me celebrated in mine; but so
far as the rest is concerned we are just two ordinary human beings,
and if we are going to be friends we must understand each other
from the first, as far as we can.'
'I'll try to do my share,' said Van Torp, taking her tone.
'Very well. I'll do mine. I began by thinking you were amusing, when
I first met you. Then you frightened me last winter, and I hated you.
Not only that, I loathed you—there's no word strong enough for
what I felt. When I saw you in the audience, you almost paralysed
my voice.'
'I didn't know it had been as bad as that,' said Mr. Van Torp quietly.
'Yes. It was worse than I can make you understand. And last spring,
when you were in so much trouble, I believed every word that was
said against you, even that you had murdered your partner's
daughter in cold blood to get rid of her, though that looked as
incredible to sensible people as it really was. It was only when I saw
how Lady Maud believed in you that I began to waver, and then I
understood.'
'I'm glad you did.'
'So am I. But she is such a good woman herself that nobody can be
really bad in whom she believes. And now I'm changed still more. I
like you, and I'm sure that we shall be friends, if you will make me
one promise and keep it.'
'What is it?'
'That you will give up all idea of ever marrying me, no matter what
happens, even if I broke——'
'It's no use to go on,' interrupted Van Torp, 'for I can't promise
anything like that. Maybe you don't realise what you're asking, but
it's the impossible. That's all.'
'Oh, nonsense!' Margaret tried to laugh lightly, but it was a failure.
'No, it's very far from nonsense,' he replied, almost sternly. 'Since
you've spoken first, I'm going to tell you several things. One is, that
I accepted the syndicate's offer for the Nickel Trust so as to be free
to take any chance that might turn up. It had been open some time,
but I accepted it on the day I heard of your engagement. That's a
big thing. Another is, that I played a regular trick on Logotheti so as
to come and see you here. I deliberately asked him to dine with me
last night in London. I went right home, wrote a note to him,
antedated for yesterday afternoon, to put him off, and I left it to be
sent at the right hour. Then I drove to the station, and here I am.
You may call that pretty sharp practice, but I believe all's fair in love
and war, and I want you to understand that I think so. There's one
thing more. I won't give up the hope of making you marry me while
you're alive and I am, not if you're an old woman, and I'll put up all
I have in the game, including my own life and other people's, if it
comes to that. Amen.'
Margaret bent her head a little and was silent.
'Now you know why I won't promise what you asked,' said Van Torp
in conclusion.
He had not raised his voice; he had not laid a heavy stress on half
his words, as he often did in common conversation; there had been
nothing dramatic in his tone; but Margaret had understood well
enough that it was the plain statement of a man who meant to
succeed, and whose strength and resources were far beyond those
of ordinary suitors. She was not exactly frightened; indeed, since her
dislike for him had melted away, it was impossible not to feel a
womanly satisfaction in the magnitude of her conquest; but she also
felt instinctively that serious trouble and danger were not far off.
'You have no right to speak like that,' she said rather weakly, after a
moment.
'Perhaps not. I don't know. But I consider that you have a right to
know the truth, and that's enough for me. It's not as if I'd made up
my mind to steal your ewe-lamb from you and put myself in its
place. Logotheti is not any sort of a ewe-lamb. He's a man, he's got
plenty of strength and determination, he's got plenty of money—
even what I choose to call plenty. He says he cares for you. All right.
So do I. He says he'll marry you. I say that I will. All right again.
You're the prize put up for the best of two fighting men. You're not
the first woman in history who's been fought for, but, by all that's
holy, there never was one better worth it, not Helen of Troy herself!'
The last few words came with a sort of stormy rush, and he turned
round suddenly, and stood with his back against the gate, thrusting
his hands deep into his coat-pockets, perhaps with the idea of
keeping them quiet; but he did not come any nearer to her, and she
felt she was perfectly safe, and that a much deeper and more lasting
power had hold of him than any mere passionate longing to take her
in his arms and press his iron lips on hers against her will. She
began to understand why he was what he was, at an age when
many successful men are still fighting for final success. He was a
crown-grasper, like John the Smith. Beside him Logotheti was but a
gifted favourite of fortune. He spoke of Helen, but if he was
comparing his rival with Paris he himself was more like an Ajax than
like good King Menelaus.
Margaret was not angry; she was hardly displeased, but she was
really at a loss what to say, and she said the first sensible thing that
suggested itself and that was approximately true.
'I'm sorry you have told me all this. We might have spent these next
two days very pleasantly together. Oh, I'm not pretending what I
don't feel! It's impossible for a woman like me, who can still be free,
not to be flattered when such a man as you cares for her in earnest,
and says the things you have. But, on the other hand, I'm engaged
to be married to another man, and it would not be loyal of me to let
you make love to me.'
'I don't mean to,' said Van Torp stoutly. 'It won't be necessary. If I
never spoke again you wouldn't forget what I've told you—ever!
Why should I say it again? I don't want to, until you can say as
much to me. If it's time to go, hitch the lead to my collar and take
me home! I'll follow you as quietly as a spaniel, anywhere!'
'And what would happen if I told you not to follow me, but to go
home and lie down in your kennel?' She laughed low as she moved
away from the gate.
'I'm not sure,' answered Van Torp. 'Don't.'
The last word was not spoken at all with an accent of warning, but it
was not said in a begging tone either. Margaret's short laugh
followed it instantly. He took the cue she offered, and went on
speaking in his ordinary manner.
'I'm not a bad dog if you don't bully me, and if you feed me at
regular hours and take me for a walk now and then. I don't pretend
I'm cut out for a French pet, because I'm not. I'm too big for a lap-
dog, and too fond of sport for the drawing-room, I suppose. A good
useful dog generally is, isn't he? Maybe I'm a little quarrelsome with
other dogs, but then, they needn't come bothering around!'
Margaret was amused, or pretended to be, but she was also thinking
very seriously of the future, and asking herself whether she ought to
send for Logotheti at once, or not. Van Torp would certainly not
leave Bayreuth at a moment's notice, at her bidding, and if he
stayed she could not now refuse to see him, with any show of
justice. She thought of a compromise, and suddenly stood still in the
lane.
'You said just now that you would not say over again any of those
things you have told me to-night. Do you mean that?'
'Yes, I mean it.'
'Then please promise that you won't. That's all I ask if you are going
to spend the next two days here, and if I am to let you see me.'
'I promise,' Van Torp answered, without hesitation.
She allowed herself the illusion that she had both done the right
thing and also taken the position of command; and he, standing
beside her, allowed himself to smile at the futility of what she was
requiring of him with so much earnestness, for little as he knew of
women's ways he was more than sure that the words he had spoken
that night would come back to her again and again; and more than
that he could not hope at present. But she could not see his face
clearly.
'Thank you,' she said. 'That shall be our compact.'
To his surprise, she held out her hand. He took it with wonderful
calmness, considering what the touch meant to him, and he
returned discreetly what was meant for a friendly pressure. She was
so well satisfied now that she did not think it necessary to telegraph
to Logotheti that he might start at once, though even if she had
done so immediately he could hardly have reached Bayreuth till the
afternoon of the next day but one, when the last performance of
Parsifal would be already going on; and she herself intended to leave
on the morning after that.
She walked forward in silence for a few moments, and the lights of
the town grew quickly brighter.
'You will come in and have some supper with us, of course,' she said
presently.
'Why, certainly, since you're so kind,' answered Van Torp.
'I feel responsible for your having forgotten to dine,' she laughed. 'I
must make it up to you. By this time Mrs. Rushmore is probably
wondering where I am.'
'Well,' said the American, 'if she thinks I'm perfection, she knows
that you're safe with me, I suppose, even if you do come home a
little late.'
'I shall say that we walked home very slowly, in order to breathe the
air.'
'Yes. We've walked home very slowly.'
'I mean,' said Margaret quickly, 'that I shall not say we have been
out towards the fields, as far as the gate.'
'I don't see any harm if we have,' observed Mr Van Torp indifferently.
'Harm? No! Don't you understand? Mrs. Rushmore is quite capable
of thinking that I have already—how shall I say?——' she stopped.
'Taken note of her good advice,' he said, completing the sentence for
her.
'Exactly! Whereas nothing could be further from my intention, as you
know. I'm very fond of Mrs. Rushmore,' Margaret continued quickly,
in order to get away from the dangerous subject she had felt obliged
to approach; 'she has been a mother to me, and heaven knows I
needed one, and she has the best and kindest heart in the world.
But she is so anxious for my happiness that, whenever she thinks it
is at stake, she rushes at conclusions without the slightest reason,
and then it's very hard to get them out of her dear old head!'
'I see. If that's why she thinks me perfection, I'll try not to
disappoint her.'
They reached the hotel, went upstairs, and separated on the landing
to get ready for supper. Margaret went to her own room, and before
joining Mrs. Rushmore she wrote a message to Alphonsine, her
theatre maid, who was visiting her family in Alsatia. Margaret
generally telegraphed her instructions, because it was much less
trouble than to write. She inquired whether Alphonsine would be
ready to join her in Paris on a certain day, and she asked for the
address of a wig-maker which she had forgotten.
On his side of the landing, Mr. Van Torp found Stemp waiting to
dress him, and the valet handed him a telegram. It was from
Captain Brown, and had been re-telegraphed from London.
'Anchored off Saint Mark's Square to-day, 3.30 P.M. Quick passage.
No stop. Coaling to-morrow. Ready for sea next morning.'
Mr. Van Torp laid the message open on the table in order to save
Stemp the trouble of looking for it afterwards.
'Stemp,' he asked, as he threw off his coat and kicked off his dusty
shoes, 'were you ever sea-sick?'
'Yes, sir,' answered the admirable valet, but he offered no more
information on the subject.
During the silence that followed, neither wasted a second. It is no
joke to wash and get into evening dress in six minutes, even with
the help of a body-servant trained to do his work at high speed.
'I mean,' said Van Torp, when he was already fastening his collar,
'are you sea-sick nowadays?'
'No, sir,' replied Stemp, in precisely the same tone as before.
'I don't mean on a twenty-thousand-ton liner. Black cravat. Yes. I
mean on a yacht. Fix it behind. Right. Would you be sea-sick on a
steam yacht?'
'No, sir.'
'Sure?'
'Yes, sir.'
'Then I'll take you. Tuxedo.'
'Thank you, sir.'
Stemp held up the dinner-jacket; Mr. Van Torp's solid arms slipped
into the sleeves, he shook his sturdy shoulders, and pulled the jacket
down in front while the valet 'settled' the back. Then he faced round
suddenly, like a soldier at drill.
'All right?' he inquired.
Stemp looked him over carefully from head to foot in the glare of the
electric light.
'Yes, sir.'
Van Torp left the room at once. He found Mrs. Rushmore slowly
moving about the supper-table, more imposing than ever in a
perfectly new black tea-gown and an extremely smart widow's cap.
Mr. Van Torp thought she was a very fine old lady indeed. Margaret
had not entered yet; a waiter with smooth yellow hair stood by a
portable sideboard on which there were covered dishes. There were
poppies and corn-flowers in a plain white jar on the table. Mrs.
Rushmore smiled at the financier; it would hardly be an
exaggeration to say that she beamed upon him. They had not met
alone since his first visit on the previous afternoon.
'Miss Donne is a little late,' she said, as if the fact were very
pleasing. 'You brought her back, of course.'
'Why, certainly,' said Mr. Van Torp with an amiable smile.
'You can hardly have come straight from the theatre,' continued the
lady, 'for I heard the other people in the hotel coming in fully twenty
minutes before you did.'
'We walked home very slowly,' said Mr. Van Torp, still smiling
amiably.
'Ah, I see! You went for a little walk to get some air!' She seemed
delighted.
'We walked home very slowly in order to breathe the air,' said Mr.
Van Torp—'to breathe the air, as you say. I have to thank you very
much for giving me your seat, Mrs. Rushmore.'
'To tell the truth,' replied the good lady, 'I was very glad to let you
take my place. I cannot say I enjoy that sort of music myself. It
gives me a headache.'
Margaret entered at this point in a marvellous 'creation' of Chinese
crape, of the most delicate shade of heliotrope. Her dressmaker
called it also a tea-gown, but Mr. Van Torp would have thought it
'quite appropriate' for a 'dinner-dance' at Bar Harbor.
'My dear child,' said Mrs. Rushmore, 'how long you were in getting
back from the theatre! I began to fear that something had
happened!'
'We walked home very slowly,' said Margaret, with a pleasant smile.
'Ah? You went for a little walk to get some air?'
'We just walked home very slowly, in order to breathe the air,'
Margaret answered innocently.
It dawned on Mr. Van Torp that the dignified Mrs. Rushmore was not
quite devoid of a sense of humour. It also occurred to him that her
repetition of the question to Margaret, and the latter's answer, must
have revealed to her the fact that the two had agreed upon what
they should say, since they used identically the same words, and
that they therefore had an understanding about something they
preferred to conceal from her. Nothing could have given Mrs.
Rushmore such profound satisfaction as this, and it revealed itself in
her bright smiles and her anxiety that both Margaret and Van Torp
should, if possible, over-eat themselves with the excellent things she
had been at pains to provide for them and for herself. For she was
something of an epicure and her dinners in Versailles were of good
fame, even in Paris.
Great appetites are generally silent, like the sincerest affections.
Margaret was very hungry, and Mr. Van Torp was both hungry and
very much in love. Mrs. Rushmore was neither, and she talked
pleasantly while tasting each delicacy with critical satisfaction.
'By the bye,' she said at last, when she saw that the millionaire was
backing his foretopsail to come to anchor, as Captain Brown might
have expressed it, 'I hope you have not had any further trouble
about your rooms, Mr. Van Torp.'
'None at all, that I know of,' answered the latter. 'My man told me
nothing.'
'The Russian prince arrived this evening while you were at the
theatre, and threatened the director with all sorts of legal
consequences because the rooms he had ordered were occupied. He
turns out to be only a count after all.'
'You don't say so,' observed Mr. Van Torp, in an encouraging tone.
'What became of him?' Margaret asked, without much interest.
'Did Potts not tell you, my dear? Why, Justine assisted at the whole
interview and came and told me at once.'
Justine was Mrs. Rushmore's Parisian maid, who always knew
everything.
'What happened?' inquired Margaret, still not much interested.
'He arrived in an automobile,' answered Mrs. Rushmore, and she
paused.
'What old Griggs calls a sudden-death-cart,' Mr. Van Torp put in.
'What a shocking name for it!' cried Mrs. Rushmore. 'And you are
always in them, my dear child!' She looked at Margaret. 'A sudden-
death-cart! It quite makes me shiver.'
'Griggs says that all his friends either kill or get killed in them,'
explained the American.
'My throat-doctor says motoring is very bad for the voice, so I've
given it up,' Margaret said.
'Really? Thank goodness your profession has been of some use to
you at last, my dear!'
Margaret laughed.
'Tell us about the Russian count,' she said. 'Has he found lodgings,
or is he going to sleep in his motor?'
'My dear, he's the most original man you ever heard of! First he
wanted to buy the hotel and turn us all out, and offered any price
for it, but the director said it was owned by a company in Munich.
Then he sent his secretary about trying to buy a house, while he
dined, but that didn't succeed either. He must be very wealthy, or
else quite mad.'
'Mad, I should say,' observed Mr. Van Torp, slowly peeling a peach.
'Did you happen to catch his name, Mrs. Rushmore?'
'Oh, yes! We heard nothing else all the afternoon. His name is
Kralinsky—Count Kralinsky.'
Mr. Van Torp continued to peel his peach scientifically and
economically, though he was aware that Margaret was looking at
him with sudden curiosity.
'Kralinsky,' he said slowly, keeping his eyes on the silver blade of the
knife as he finished what he was doing. 'It's not an uncommon
name, I believe. I've heard it before. Sounds Polish, doesn't it?'
He looked up suddenly and showed Margaret the peeled peach on
his fork. He smiled as he met her eyes, and she nodded so slightly
that Mrs. Rushmore did not notice the movement.
'Did you ever see that done better?' he asked with an air of triumph.
'Ripping!' Margaret answered. 'You're a dandy dab at it!'
'My dear child, what terrible slang!'
'I'm sorry,' said Margaret. 'I'm catching all sorts of American
expressions from Mr. Van Torp, and when they get mixed up with my
English ones the result is Babel, I suppose!'
'I've not heard Mr. Van Torp use any slang expressions yet, my dear,'
said Mrs. Rushmore, almost severely.
'You will,' Margaret retorted with a laugh. 'What became of Count
Kralinsky? I didn't mean to spoil your story.'
'My dear, he's got the Pastor to give up his house, by offering him a
hundred pounds for the poor here.'
'It's cheap,' observed Mr. Van Torp. 'The poor always are.'
'You two are saying the most dreadful things to-night!' cried Mrs.
Rushmore.
'Nothing dreadful in that, Mrs. Rushmore,' objected the millionaire.
'There's no investment on earth like charity.'
'We are taught that by charity we lay up treasures in heaven,' said
the good lady.
'Provided it's not mentioned in the newspapers,' retorted Mr. Van
Torp. 'When it is, we lay up treasures on earth. I don't like to
mention other men in that connexion, especially as I've done the
same thing myself now and then, just to quiet things down; but I
suppose some names will occur to you right away, don't they?
Where is the Pastor going to sleep, now that the philanthropist has
bought him out?'
'I really don't know,' answered Mrs. Rushmore.
'Then he's the real philanthropist,' said Van Torp. 'If he understood
the power of advertisement, and wanted it, he'd let it be known that
he was going to sleep on the church steps without enough blankets,
for the good of the poor who are to have the money, and he'd get
everybody to come and look at him in his sleep, and notice how
good he was. Instead of that, he's probably turned in under the back
stairs, in the coal-hole, without saying anything about it. I don't
know how it strikes you, Mrs. Rushmore, but it does seem to me
that the clergyman's the real philanthropist after all!'
'Indeed he is, poor man,' said Margaret, a good deal surprised at
Van Torp's sermon on charity, and wondering vaguely whether he
was talking for effect or merely saying what he really thought.
An effect certainly followed.
'You put it very sensibly, I'm sure,' said Mrs. Rushmore, 'though of
course I should not have looked for anything else from a fellow-
countryman I respect. You startled me a little at first, when you said
that the poor are always cheap! Only that, I assure you.'
'Well,' answered the American, 'I never was very good at expressing
myself, but I'm glad we think alike, for I must say I value your
opinion very highly, Mrs. Rushmore, as I had learned to value the
opinion of your late husband.'
'You're very kind,' she said, in a grateful tone.
Margaret was not sure that she was pleased as she realised how
easily Van Torp played upon her old friend's feelings and convictions,
and she wondered whether he had not already played on her own
that night, in much the same way. But with the mere thought his
words and his voice came back to her, with his talk about the
uselessness of ever repeating what he had said that once, because
he knew she could never forget it. And her young instinct told her
that he dealt with the elderly woman precisely as if she were a man,
with all the ease that proceeded from his great knowledge of men
and their weaknesses; but that with herself, in his ignorance of
feminine ways, he could only be quite natural.
He left them soon after supper, and gave himself up to Stemp,
pondering over what he had accomplished in two days, and also
about another question which had lately presented itself. When he
was ready to send his valet to bed he sat down at his table and
wrote a telegram:
'If you can find Barak, please explain that I was mistaken. Kralinsky
is not in New York, but here in Bayreuth for some days, lodging at
the Pastor's house.'
This message was addressed to Logotheti at his lodgings in London,
and Van Torp signed it and gave it to Stemp to be sent at once.
Logotheti never went to bed before two o'clock, as he knew, and
might very possibly get the telegram the same night.
When his man was gone, Van Torp drew his chair to the open
window and sat up a long time thinking about what he had just
done; for though he held that all was fair in such a contest, he did
not mean to do anything which he himself thought 'low-down.' One
proof of this odd sort of integrity was that the telegram itself was a
fair warning of his presence in Bayreuth, where Logotheti knew that
Margaret was still stopping.
As for the rest, he was quite convinced that it was Kralinsky himself,
the ruby merchant, who had suddenly appeared at Bayreuth, and
that this man was no other than the youth he had met long ago as a
cow-boy in the West, who used to whistle Parsifal with his
companion in exile, and who, having grown rich, had lost no time in
coming to Europe for the very purpose of hearing the music he had
always loved so well. And that this man had robbed the poor Tartar
girl, Mr. Van Torp had no manner of doubt; and he believed that he
had probably promised her marriage and abandoned her; and if this
were true, to help her to find Kralinsky was in itself a good action.
 CHAPTER VII

When Van Torp and Logotheti left Mr. Pinney's shop, the old jeweller
meant to have a good look at the ruby the Greek had brought him,
and was going to weigh it, not merely as a matter of business, for
he weighed every stone that passed through his hands from crown
diamonds to sparks, but with genuine curiosity, because in a long
experience he had not seen very many rubies of such a size, which
were also of such fine quality, and he wondered where this one had
been found.
Just then, however, two well-dressed young men entered the shop
and came up to him. He had never seen either of them before, but
their looks inspired him with confidence; and when they spoke, their
tone was that of English gentlemen, which all other Englishmen find
it practically impossible to imitate, and which had been extremely
familiar to Mr. Pinney from his youth. Though he was the great
jeweller himself, the wealthy descendant of five of his name in
succession, and much better off than half his customers, he was
alone in his shop that morning. The truth was that his only son, the
sixth Pinney and the apple of his eye, had just been married and
was gone abroad for a honeymoon trip, and the head shopman, who
was Scotch, was having his month's holiday in Ayrshire, and the
second man had been sent for, to clean and restring the Duchess of
Barchester's pearls at her Grace's house in Cadogan Gardens, as was
always done after the season, and a couple of skilled workmen for
whom Mr. Pinney found occupation all the year round were in the
workshop at their tables; wherefore, out of four responsible and
worthy men who usually were about, only the great Mr. Pinney
himself was at his post.
One of the two well-dressed customers asked to see some pins, and
the other gave his advice. The first bought a pin with a small
sapphire set in sparks for ten guineas, and gave only ten pounds for
it because he paid cash. Mr. Pinney put the pin into its little morocco
case, wrapped it up neatly and handed it to the purchaser. The latter
and his friend said good-morning in a civil and leisurely manner,
sauntered out, took a hansom a few steps farther down the street,
and drove away.
The little paper twist containing Logotheti's ruby was still exactly
where Mr. Pinney had placed it on the counter, and he was going to
examine the stone and weigh it at last, when two more customers
entered the shop, evidently foreigners, and moreover of a sort
unfamiliar to the good jeweller, and especially suspicious.
The two were Baraka and her interpreter and servant, whom
Logotheti had called a Turk, and who was really a Turkish subject
and a Mohammedan, though as to race, he was a half-bred Greek
and Dalmatian. Now Dalmatians are generally honest, truthful, and
trustworthy, and the low-class Greek of Constantinople is usually
extremely sharp, if he is nothing more definitely reprehensible; and
Baraka's man was a cross between the two, as I have said, and had
been brought up as a Musulman in a rich Turkish family, and
recommended to Baraka by the Persian merchant in whose house
she had lived. He had been originally baptized a Christian under the
name of Spiro, and had been subsequently renamed Selim when he
was made a real Moslem at twelve years old; so he used whichever
name suited the circumstances in which he was placed. At present
he was Spiro. He was neatly dressed in grey clothes made by a
French tailor, and he wore a French hat, which always made a bad
impression on Mr. Pinney. He had brown hair, brown eyes, a brown
moustache, and a brown face; he looked as active as a cat, and Mr.
Pinney at once put him down in his mind as a 'Froggy.' But the
jeweller was less sure about Baraka, who was dressed like any
young Englishman, but looked like no European he had ever seen.
On the whole, he took the newcomer for the son of an Indian rajah
sent to England to be educated.
The interpreter spoke broken but intelligible English. He called
Baraka his master, and explained that the latter wished to see some
rubies, if Mr. Pinney had any, cut or uncut. The young gentleman, he
said, did not speak English, but was a good judge of stones.
For one moment the jeweller forgot the little paper twist as he
turned towards his safe, pulling out his keys at the same time. To
reach the safe he had to walk the whole length of the shop, behind
the counter, and before he had gone half way he remembered the
stone, turned, came back, and slipped it into his waistcoat pocket.
Then he went and got the little japanned strong-box with a patent
lock, in which he kept loose stones, some wrapped up in little pieces
of paper, and some in pill-boxes. He brought it to his customers, and
opened it before them.
They stayed a long time, and Spiro asked many questions for
Baraka, chiefly relating to the sliding-scale of prices which is
regulated by the weight of the stones where their quality is equally
good, and Baraka made notes of some sort in a little English
memorandum-book, as if she had done it all her life; but Mr. Pinney
could not see what she wrote. He was very careful, and watched the
stones, when she took them in her fingers and held them up against
the light, or laid them on a sheet of white paper to look at them
critically.
She bought nothing; and when she had seen all he had to show her,
she thanked him very much through Spiro, said she would come
back another day, and went out with a leisurely, Oriental gait, as if
nothing in the world could hurry her. Mr. Pinney counted the stones
again, and was going to lock the box, when his second man came in,
having finished stringing the Duchess's pearls. At the same moment,
it occurred to Mr. Pinney that he might as well go to luncheon, and
that he had better put Logotheti's ruby into the little strong-box and
lock it up in the safe until he at last had a chance to weigh it. He
accordingly took the screw of paper from his waistcoat pocket, and
as a matter of formality he undid it once more.
'Merciful Providence!' cried Mr. Pinney, for he was a religious man.
The screw of paper contained a bit of broken green glass. He threw
his keys to his shopman without another word, and rushed out into
the street without his hat, his keen old face deadly pale, and his
beautiful frock-coat flying in his wake.
He almost hurled himself upon a quiet policeman.
'Thief!' he cried. 'Two foreigners in grey clothes—ruby worth ten
thousand pounds just gone—I'm Pinney the jeweller!'
You cannot astonish a London policeman. The one Pinney had
caught looked quietly up and down the street, and then glanced at
his interlocutor to be sure that it was he, for he knew him by sight.
'All right,' he said quickly, but very quietly. 'I'll have them in a
minute, sir, for they're in sight still. Better go in while I take them,
sir.'
He caught them in less than a minute without the slightest difficulty,
and by some odd coincidence two other policemen suddenly
appeared quite close to him. There was a little stir in the street, but
Baraka and Spiro were too sensible and too sure of themselves to
offer any useless resistance, and supposing there was some
misunderstanding they walked back quietly to Mr. Pinney's shop
between two of the policemen, while the third went for a four-
wheeler at the nearest stand, which happened to be the corner of
Brook Street and New Bond Street.
Mr. Pinney recognised his late customers without hesitation, and
went with them to the police station, where he told his story and
showed the piece of green glass. Spiro tried to speak, but was
ordered to hold his tongue, and as no rubies were found in their
pockets he and Baraka were led away to be more thoroughly
searched.
But now, at last, Baraka resisted, and with such tremendous energy
that there would have been serious trouble if Spiro had not called
out something which at once changed the aspect of matters.
'Master is lady!' he yelled. 'Lady, man clothes!'
'That makes a pretty bad case,' observed the sergeant who was
superintending. 'Send for Mrs. Mowle.'
Baraka did not resist when she saw the matron, and went quietly
with her to a cell at the back of the station. In less than ten minutes
Mrs. Mowle came out and locked the door after her. She was a
cheery little person, very neatly dressed, and she had restless bright
eyes like a ferret. She brought a little bag of soft deerskin in her
hand, and a steel bodkin with a wrought silver handle, such as
southern Italian women used to wear in their hair before such
weapons were prohibited. Mrs. Mowle gave both objects to the
officer without comment.
'Any scars or tattoo-marks, Mrs. Mowle?' he inquired in his business-
like way.
'Not a one,' answered Mrs. Mowle, who had formerly taken in
washing at home and was the widow of a brave policeman, killed in
doing his duty.
In the bag there were several screws of paper, which were found to
contain uncut rubies of different sizes to a large value. But there was
one, much larger than the others, which Mr. Van Torp had not seen
that morning. Mr. Pinney looked at it very carefully, held it to the
light, laid it on a sheet of paper, and examined it long in every
aspect. He was a conscientious man.
'To the best of my belief,' he deposed, 'this is the stone that was on
my counter half an hour ago, and for which this piece of green glass
was substituted. It is the property of a customer of mine, Monsieur
Konstantin Logotheti of Paris, who brought it to me this morning to
be cut. I think it may be worth between nine and ten thousand
pounds. I can say nothing as to the identity of the paper, for tissue
paper is very much alike everywhere.'
'The woman,' observed the officer in charge of the station, 'appears
to steal nothing but rubies. It looks like a queer case. We'll lock up
the two, Mr. Pinney, and if you will be kind enough to look in to-
morrow morning, I'm sure the Magistrate won't keep you waiting for
the case.'
Vastly relieved and comforted, Mr. Pinney returned to his shop.
Formality required that the ruby itself, with the others in the bag,
should remain in the keeping of the police till the Magistrate ordered
it to be returned to its rightful owner, the next morning; but Mr.
Pinney felt quite as sure of its safety as if it were in the japanned
strong-box in his own safe, and possibly even a little more sure, for
nobody could steal it from the police station.
But after he was gone, Spiro was heard calling loudly, though not
rudely or violently, from his place of confinement.
'Mr. Policeman! Mr. Policeman! Please come speak!'
The man on duty went to the door and asked what he wanted. In
his broken English he explained very clearly that Baraka had a friend
in London who was one of the great of the earth, and who would
certainly prove her innocence, vouch for her character, and cause
her to be set at large without delay, if he knew of her trouble.
'What is the gentleman's name?' inquired the policeman.
The name of Baraka's friend was Konstantin Logotheti, and Spiro
knew the address of the lodgings he always kept in St. James's
Place.
'Very well,' said the policeman. 'I'll speak to the officer at once.'
'I thank very much, sir,' Spiro answered, and he made no more
noise.
The sergeant looked surprised when the message was given to him.
'Queer case this,' he observed. 'Here's the thief appealing to the
owner of the stolen property for help; and the owner is one of those
millionaire financiers; and the thief is a lovely girl in man's clothes.
By the bye, Sampson, tell Mrs. Mowle to get out some women's
slops and dress her decently, while I see if I can find Mr. Logotheti
by telephone. They'll be likely to know something about him at the
Bank if he's not at home, and he may come to find out what's the
matter. If Mrs. Foxwell should look in and want to see the girl, let
her in, of course, without asking me. If she's in town, she'll be here
before long, for I've telephoned to her house, as usual when there's
a girl in trouble.'
There was a sort of standing, unofficial order that in any case of a
girl or a young woman being locked up, Mrs. Foxwell was to know of
it, and she had a way of remembering a great many sergeants'
names, and doing kind things for their wives at Christmas-time,
which further disposed them to help her in her work. But the London
police are by nature the kindliest set of men who keep order
anywhere in the world, and they will readily help a man or woman
who tries to do good in a sensible, practical way; and if they are
sometimes a little prejudiced in favour of their own perspicuity in
getting up a case, let that policeman, of any other country, who is
quite without fault, throw the first stone at their brave, good natured
heads.
Logotheti was not at his lodgings in St. James's Place, and from each
of two clubs to which the officer telephoned rather at random, the
only answer was that he was a member but not in the house. The
officer wrote a line to his rooms and sent it by a messenger, to be
given to him as soon as he came in.
 "She grasped Lady Maud's hand."

It was late in the hot afternoon when Mrs. Foxwell answered the
message by coming to the police station herself. She was at once
admitted to Baraka's cell and the door was closed after her.
The girl was lying on the pallet bed, dressed in a poor calico skirt
and a loose white cotton jacket, which Mrs. Mowle had brought and
had insisted that she must put on; and her man's clothes had been
taken from her with all her other belongings. She sat up, forlorn,
pale and lovely, as the kind visitor entered and stood beside her.
'Poor child!' exclaimed the lady, touched by her sad eyes. 'What can
I do to help you?'
Baraka shook her head, for she did not understand. Then she looked
up into eyes almost as beautiful as her own, and pronounced a
name, slowly and so distinctly that it was impossible not to hear
each syllable.
'Konstantin Logotheti.'
The lady started, as well she might; for she was no other than Lady
Maud, who called herself by her own family name, 'Mrs. Foxwell,' in
her work amongst the poor women of London.
Baraka saw the quick movement and understood that Logotheti was
well known to her visitor. She grasped Lady Maud's arm with both
her small hands, and looked up to her face with a beseeching look
that could not be misunderstood. She wished Logotheti to be
informed of her captivity, and was absolutely confident that he
would help her out of her trouble. Lady Maud was less sure of that,
however, and said so, but it was soon clear that Baraka did not
speak a word of any language known to Lady Maud, who was no
great linguist at best. Under these circumstances it looked as if there
were nothing to be done for the poor girl, who made all sorts of
signs of distress, when she saw that the English woman was about
to leave her, in sheer despair of being of any use. Just then,
however, the sergeant came to the door, and informed the visitor
that the girl had an accomplice who spoke her language and knew
some English, and that by stretching a point he would bring the
man, if Mrs. Foxwell wished to talk with him.
The result was that in less than half an hour, Lady Maud heard from
Spiro a most extraordinary tale, of which she did not believe a single
word. To her plain English mind, it all seemed perfectly mad at first,
and on reflection she thought it an outrageous attempt to play upon
her credulity; whereas she was thoroughly convinced that the girl
had come to grief in some way through Logotheti and had followed
him from Constantinople, probably supporting herself and her
companion by stealing on the way. Lady Maud's husband had been a
brute, but he knew the East tolerably well, having done some
military duty in the Caucasus before he entered the diplomatic
service; his stories had chiefly illustrated the profound duplicity of all
Asiatics, and she had not seen any reason to disbelieve them.
When Spiro had nothing more to say, therefore, she rose from the
only seat there was and shook her head with an air of utter
incredulity, mingled with the sort of pitying contempt she felt for all
lying in general. She could easily follow the case, by the help of the
sergeant and the Police Court reports, and she might be able to help
Baraka hereafter when the girl had served the sentence she would
certainly get for such an important and cleverly managed theft. The
poor girl implored and wept in vain; Lady Maud could do nothing,
and would not stay to be told any more inane stories about ruby
mines in Tartary. She called the sergeant, freed herself from Baraka's
despairing hold on her hand and went out. Spiro was then marched
back to his cell on the men's side.
Though it was hot, Lady Maud walked home, as Mr. Van Torp had
done that same morning when he had left Mr. Pinney's shop. She
always walked when she was in any distress or difficulty, for the
motion helped her to think, since she was strong and healthy, and
only in her twenty-ninth year. Just now, too, she was a good deal
disturbed by what had happened, besides being annoyed by the
attempt that had been made to play on her credulity in such a gross
way.
She was really fond of Margaret Donne, quite apart from any
admiration she felt for the Primadonna's genius, by which she might
have been influenced. In her opinion, the Tartar girl's appeal for help
to reach Logotheti could only mean one thing, and that was very far
from being to his credit. If the girl had not been positively proved to
be a thief and if she had not attempted to impose upon her by what
seemed the most absurd falsehoods, Lady Maud would very
probably have taken her under her own protection, as far as the law
would allow. But her especial charity was not for criminals or cheats,
though she had sometimes helped and comforted women accused of
far worse crimes than stealing. In this instance she could do nothing,
and she did not even wish to do anything. It was a flagrant case,
and the law would deal with it in the right way. The girl had come to
grief, no doubt, by trusting Logotheti blindly, and he had thrown her
off; if she had sunk into the dismal depths of woe behind the Virtue-
Curtain, as most of her kind did, Lady Maud would have gone in and
tried to drag her out, as she had saved others. But Logotheti's victim
had taken a different turn, had turned thief and had got into the
hands of justice. Her sin would be on his head, no doubt, but no
power could avert from her the just consequences of a misdeed that
had no necessary connexion with her fall.
Thus argued Lady Maud, while Baraka lay on her pallet bed in her
calico skirt and white cotton jacket, neither weeping, nor despairing
by any means, nor otherwise yielding to girlish weakness, but
already devising means for carrying on her pursuit of the man she
would still seek, even throughout the whole world, though she was
just now a penniless girl locked up as a thief in a London police
station. It was not one of the down-hearted, crying sort that could
have got so far already, against such portentous odds.
She guessed well enough that she would be tried the next morning
in the Police Court; for Spiro, who knew much about Europe, and
England in particular, had told her a great deal during their travels.
She had learned that England was a land of justice, and she would
probably get it in the end; for the rest, she was a good Musulman
girl and looked on whatsoever befell her as being her portion, for
good or evil, to be accepted without murmuring.
Lady Maud could not know anything of this and took Baraka for a
common delinquent, so far as her present situation was concerned.
But when the Englishwoman thought of what must have gone
before, and of the part Logotheti had almost certainly played in the
girl's life, her anger was roused, and she sat down and wrote to
Margaret on the impulse of the moment. She gave a detailed
account of her experience at the police station, including especially a
description of the way Baraka had behaved in trying to send a
message to Logotheti.
Welcome to our website – the ideal destination for book lovers and
knowledge seekers. With a mission to inspire endlessly, we offer a
vast collection of books, ranging from classic literary works to
specialized publications, self-development books, and children's
literature. Each book is a new journey of discovery, expanding
knowledge and enriching the soul of the reade

Our website is not just a platform for buying books, but a bridge
connecting readers to the timeless values of culture and wisdom. With
an elegant, user-friendly interface and an intelligent search system,
we are committed to providing a quick and convenient shopping
experience. Additionally, our special promotions and home delivery
services ensure that you save time and fully enjoy the joy of reading.

Let us accompany you on the journey of exploring knowledge and

personal growth!

textbookfull.com