100% found this document useful (4 votes)
13 views

Get Foundations of Data Science with Python 1st Edition John M. Shea PDF ebook with Full Chapters Now

Python

Uploaded by

essaiepyles
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
100% found this document useful (4 votes)
13 views

Get Foundations of Data Science with Python 1st Edition John M. Shea PDF ebook with Full Chapters Now

Python

Uploaded by

essaiepyles
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 61

Download Full ebookname - Read Now at ebookname.

com

Foundations of Data Science with Python 1st


Edition John M. Shea

https://ebookname.com/product/foundations-of-data-science-
with-python-1st-edition-john-m-shea/

OR CLICK BUTTON

DOWLOAD EBOOK

Discover More Ebook - Explore Now at ebookname.com


Instant digital products (PDF, ePub, MOBI) available
Download now and explore formats that suit you...

Data Science from Scratch First Principles with Python 1st


Edition Joel Grus

https://ebookname.com/product/data-science-from-scratch-first-
principles-with-python-1st-edition-joel-grus/

ebookname.com

Managing Your Biological Data with Python 1st Edition Via

https://ebookname.com/product/managing-your-biological-data-with-
python-1st-edition-via/

ebookname.com

Data Visualization with Python and JavaScript Scrape Clean


Explore Transform Your Data 1st Edition Kyran Dale

https://ebookname.com/product/data-visualization-with-python-and-
javascript-scrape-clean-explore-transform-your-data-1st-edition-kyran-
dale/
ebookname.com

The Healthy Body Handbook A Total Guide to the Prevention


and Treatment of Sports Injuries 1st Edition David Saidoff

https://ebookname.com/product/the-healthy-body-handbook-a-total-guide-
to-the-prevention-and-treatment-of-sports-injuries-1st-edition-david-
saidoff/
ebookname.com
The Pursuit of Excellence Through Education 1st Edition
Michel Ferrari

https://ebookname.com/product/the-pursuit-of-excellence-through-
education-1st-edition-michel-ferrari/

ebookname.com

Not Quite the Diplomat Home Truths About World Affairs


First Edition Chris Patten

https://ebookname.com/product/not-quite-the-diplomat-home-truths-
about-world-affairs-first-edition-chris-patten/

ebookname.com

Nothing to Be Frightened Of Julian Barnes

https://ebookname.com/product/nothing-to-be-frightened-of-julian-
barnes/

ebookname.com

Grassroots Coalitions and State Policy Change Organizing


for Immigrant Health Care 1st Edition Margaret A. Post

https://ebookname.com/product/grassroots-coalitions-and-state-policy-
change-organizing-for-immigrant-health-care-1st-edition-margaret-a-
post/
ebookname.com

Hyperbolic Dynamics and Brownian Motion An Introduction


Jacques Franchi

https://ebookname.com/product/hyperbolic-dynamics-and-brownian-motion-
an-introduction-jacques-franchi/

ebookname.com
Creating Christian Granada Society and Religious Culture
in an Old World Frontier City 1492 1600 1st Edition David
Coleman
https://ebookname.com/product/creating-christian-granada-society-and-
religious-culture-in-an-old-world-frontier-city-1492-1600-1st-edition-
david-coleman/
ebookname.com
Foundations of Data
Science with Python

Foundations of Data Science with Python introduces readers to the fundamentals of data science, includ-
ing data manipulation and visualization, probability, statistics, and dimensionality reduction. This book
is targeted toward engineers and scientists, but it should be readily understandable to anyone who knows
basic calculus and the essentials of computer programming. It uses a computational-first approach to data
science: the reader will learn how to use Python and the associated data-science libraries to visualize,
transform, and model data, as well as how to conduct statistical tests using real data sets. Rather than rely-
ing on obscure formulas that only apply to very specific statistical tests, this book teaches readers how to
perform statistical tests via resampling; this is a simple and general approach to conducting statistical tests
using simulations that draw samples from the data being analyzed. The statistical techniques and tools are
explained and demonstrated using a diverse collection of data sets to conduct statistical tests related to
contemporary topics, from the effects of socioeconomic factors on the spread of the COVID-19 virus to
the impact of state laws on firearms mortality.

This book can be used as an undergraduate textbook for an Introduction to Data Science course or to
provide a more contemporary approach in courses like Engineering Statistics. However, it is also intended
to be accessible to practicing engineers and scientists who need to gain foundational knowledge of data
science.

Key Features:
• Applies a modern, computational approach to working with data
• Uses real data sets to conduct statistical tests that address a diverse set of contemporary issues
• Teaches the fundamentals of some of the most important tools in the Python data-science stack
• Provides a basic, but rigorous, introduction to Probability and its application to Statistics
• Offers an accompanying website that provides a unique set of online, interactive tools to help the
reader learn the material
John M. Shea, PhD is a Professor in the Department of Electrical and Computer Engineering at the Uni-
versity of Florida, where he has taught classes on stochastic methods, data science, and wireless com-
munications for over 20 years. He earned his PhD in Electrical Engineering from Clemson University in
1998 and later received the Outstanding Young Alumni award from the Clemson College of Engineering
and Science. Dr. Shea was co-leader of Team GatorWings, which won the Defense Advanced Research
Project Agency’s (DARPA’s) Spectrum Collaboration Challenge (DARPA’s fifth Grand Challenge) in 2019.
He received the Lifetime Achievement Award for Technical Achievement from the IEEE Military Com-
munications Conference (MILCOM) and is a two-time winner of the Ellersick Award from the IEEE Com-
munications Society for the Best Paper in the Unclassified Program of MILCOM. He has been an editor
for IEEE Transactions on Wireless Communications, IEEE Wireless Communications magazine, and IEEE
Transactions on Vehicular Technology.
Chapman & Hall/CRC
The Python Series
About the Series

Python has been ranked as the most popular programming language, and it is widely used in education and
industry. This book series will offer a wide range of books on Python for students and professionals. Titles
in the series will help users learn the language at an introductory and advanced level, and explore its many
applications in data science, AI, and machine learning. Series titles can also be supplemented with Jupyter
notebooks.

Image Processing and Acquisition using Python, Second Edition


Ravishankar Chityala, Sridevi Pudipeddi

Python Packages
Tomas Beuzen and Tiffany-Anne Timbers

Statistics and Data Visualisation with Python


Jesús Rogel-Salazar

Introduction to Python for Humanists


William J.B. Mattingly

Python for Scientific Computation and Artificial Intelligence


Stephen Lynch

Learning Professional Python Volume 1: The Basics


Usharani Bhimavarapu and Jude D. Hemanth

Learning Professional Python Volume 2: Advanced


Usharani Bhimavarapu and Jude D. Hemanth

Learning Advanced Python from Open Source Projects


Rongpeng Li

Foundations of Data Science with Python


John Mark Shea

For more information about this series please visit: https://www.crcpress.com/Chapman--HallCRC/book-


series/PYTH
Foundations of Data
Science with Python

John M. Shea
Designed cover image: © Agnes Shea

MATLAB is a trademark of The MathWorks, Inc. and is used with permission. The MathWorks does
not warrant the accuracy of the text or exercises in this book. This book’s use or discussion of MATLAB
software or related products does not constitute endorsement or sponsorship by The MathWorks of a
particular pedagogical approach or particular use of the MATLAB software.
First edition published 2024
by CRC Press
2385 NW Executive Center Drive, Suite 320, Boca Raton FL 33431

and by CRC Press


4 Park Square, Milton Park, Abingdon, Oxon, OX14 4RN

CRC Press is an imprint of Taylor & Francis Group, LLC

© 2024 John Mark Shea

Reasonable efforts have been made to publish reliable data and information, but the author and publisher
cannot assume responsibility for the validity of all materials or the consequences of their use. The authors
and publishers have attempted to trace the copyright holders of all material reproduced in this publica-
tion and apologize to copyright holders if permission to publish in this form has not been obtained. If any
copyright material has not been acknowledged please write and let us know so we may rectify in any future
reprint.

Except as permitted under U.S. Copyright Law, no part of this book may be reprinted, reproduced, trans-
mitted, or utilized in any form by any electronic, mechanical, or other means, now known or hereafter
invented, including photocopying, microfilming, and recording, or in any information storage or retrieval
system, without written permission from the publishers.

For permission to photocopy or use material electronically from this work, access www.copyright.com or
contact the Copyright Clearance Center, Inc. (CCC), 222 Rosewood Drive, Danvers, MA 01923, 978-750-
8400. For works that are not available on CCC please contact [email protected]

Trademark notice: Product or corporate names may be trademarks or registered trademarks and are used
only for identification and explanation without intent to infringe.

ISBN: 978-1-032-34674-8 (hbk)


ISBN: 978-1-032-35042-4 (pbk)
ISBN: 978-1-003-32499-7 (ebk)

DOI: 10.1201/9781003324997

Typeset in Latin Modern font


by KnowledgeWorks Global Ltd.

Publisher’s note: This book has been prepared from camera-ready copy provided by the authors.
For Tucker, Charlotte, and Amelia—

proof that events with zero probability


(having three children that are this wonderful)
happen!
Taylor & Francis
Taylor & Francis Group
http://taylorandfrancis.com
Contents

Acknowledgments xi

Preface xiii

1 Introduction 1
1.1 Who is this book for? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Why learn data science from this book? . . . . . . . . . . . . . . . . . . . . 1
1.3 What is data science? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.4 What data science topics does this book cover? . . . . . . . . . . . . . . . . 5
1.5 What data science topics does this book not cover? . . . . . . . . . . . . . 6
1.6 Extremely Brief Introduction to Jupyter and Python . . . . . . . . . . . . . 6
1.7 Chapter Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16

2 First Simulations, Visualizations, and Statistical Tests 17


2.1 Motivating Problem: Is This Coin Fair? . . . . . . . . . . . . . . . . . . . . 17
2.2 First Computer Simulations . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
2.3 First Visualizations: Scatter Plots and Histograms . . . . . . . . . . . . . . 20
2.4 First Statistical Tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
2.5 Chapter Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31

3 First Visualizations and Statistical Tests with Real Data 33


3.1 Introduction to Pandas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
3.2 Visualizing Multiple Data Sets – Part 1: Scatter Plots . . . . . . . . . . . . 39
3.3 Partitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
3.4 Summary Statistics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
3.5 Visualizing Multiple Data Sets – Part 2: Histograms for Partitioned Data . 56
3.6 Null Hypothesis Testing with Real Data . . . . . . . . . . . . . . . . . . . . 61
3.7 A Quick Preview of Two-Dimensional Statistical Methods . . . . . . . . . . 73
3.8 Chapter Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76

4 Introduction to Probability 77
4.1 Outcomes, Sample Spaces, and Events . . . . . . . . . . . . . . . . . . . . . 77
4.2 Relative Frequencies and Probabilities . . . . . . . . . . . . . . . . . . . . . 78
4.3 Fair Experiments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
4.4 Axiomatic Probability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
4.5 Corollaries to the Axioms of Probability . . . . . . . . . . . . . . . . . . . . 94
4.6 Combinatorics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
4.7 Chapter Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112

vii
viii Contents

5 Null Hypothesis Tests 113


5.1 Statistical Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113
5.2 General Resampling Approaches for Null Hypothesis Significance Testing . 120
5.3 Calculating p-Values . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
5.4 How to Sample from the Pooled Data . . . . . . . . . . . . . . . . . . . . . 132
5.5 Example Null Hypothesis Significance Tests . . . . . . . . . . . . . . . . . . 138
5.6 Bootstrap Distribution and Confidence Intervals . . . . . . . . . . . . . . . 141
5.7 Types of Errors and Statistical Power . . . . . . . . . . . . . . . . . . . . . 147
5.8 Chapter Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150

6 Conditional Probability, Dependence, and Independence 151


6.1 Simulating and Counting Conditional Probabilities . . . . . . . . . . . . . . 152
6.2 Conditional Probability: Notation and Intuition . . . . . . . . . . . . . . . . 157
6.3 Formally Defining Conditional Probability . . . . . . . . . . . . . . . . . . . 159
6.4 Relating Conditional and Unconditional Probabilities . . . . . . . . . . . . . 162
6.5 More on Simulating Conditional Probabilities . . . . . . . . . . . . . . . . . 164
6.6 Statistical Independence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 166
6.7 Conditional Probabilities and Independence in Fair Experiments . . . . . . 172
6.8 Conditioning and (In)dependence . . . . . . . . . . . . . . . . . . . . . . . . 176
6.9 Chain Rules and Total Probability . . . . . . . . . . . . . . . . . . . . . . . 178
6.10 Chapter Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 186

7 Introduction to Bayesian Methods 187


7.1 Bayes’ Rule . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187
7.2 Bayes’ Rule in Systems with Hidden State . . . . . . . . . . . . . . . . . . . 195
7.3 Optimal Decisions for Discrete Stochastic Systems . . . . . . . . . . . . . . 197
7.4 Bayesian Hypothesis Testing . . . . . . . . . . . . . . . . . . . . . . . . . . . 206
7.5 Chapter Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215

8 Random Variables 217


8.1 Definition of a Real Random Variable . . . . . . . . . . . . . . . . . . . . . 217
8.2 Discrete Random Variables . . . . . . . . . . . . . . . . . . . . . . . . . . . 229
8.3 Cumulative Distribution Functions . . . . . . . . . . . . . . . . . . . . . . . 237
8.4 Important Discrete RVs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 247
8.5 Continuous Random Variables . . . . . . . . . . . . . . . . . . . . . . . . . . 265
8.6 Important Continuous Random Variables . . . . . . . . . . . . . . . . . . . 279
8.7 Histograms of Continuous Random Variables and Kernel Density Estimation 299
8.8 Conditioning with Random Variables . . . . . . . . . . . . . . . . . . . . . . 301
8.9 Chapter Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 305

9 Expected Value, Parameter Estimation, and Hypothesis Tests on


Sample Means 306
9.1 Expected Value . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 306
9.2 Expected Value of a Continuous Random Variable with SymPy . . . . . . . 312
9.3 Moments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 315
9.4 Parameter Estimation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 326
9.5 Confidence Intervals for Estimates . . . . . . . . . . . . . . . . . . . . . . . 333
9.6 Testing a Difference of Means . . . . . . . . . . . . . . . . . . . . . . . . . . 342
9.7 Sampling and Bootstrap Distributions of Parameters . . . . . . . . . . . . . 356
9.8 Effect Size, Power, and Sample Size Selection . . . . . . . . . . . . . . . . . 360
9.9 Chapter Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 362
Contents ix

10 Decision-Making with Observations from Continuous Distributions 364


10.1 Binary Decisions from Continuous Data: Non-Bayesian Approaches . . . . . 364
10.2 Point Conditioning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 375
10.3 Optimal Bayesian Decision-Making with Continuous Random Variables . . 380
10.4 Chapter Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 386

11 Categorical Data, Tests for Dependence, and Goodness of Fit for


Discrete Distributions 387
11.1 Tabulating Categorical Data and Creating a Test Statistic . . . . . . . . . . 388
11.2 Null Hypothesis Significance Testing for Dependence in Contingency Tables 394
11.3 Chi-Square Goodness-of-Fit Test . . . . . . . . . . . . . . . . . . . . . . . . 400
11.4 Chapter Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 408

12 Multidimensional Data: Vector Moments and Linear Regression 409


12.1 Summary Statistics for Vector Data . . . . . . . . . . . . . . . . . . . . . . 410
12.2 Linear Regression . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 422
12.3 Null Hypothesis Tests for Correlation . . . . . . . . . . . . . . . . . . . . . . 435
12.4 Nonlinear Regression Tests . . . . . . . . . . . . . . . . . . . . . . . . . . . 437
12.5 Chapter Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 445

13 Working with Dependent Data in Multiple Dimensions 446


13.1 Jointly Distributed Pairs of Random Variables . . . . . . . . . . . . . . . . 446
13.2 Standardization and Linear Transformations . . . . . . . . . . . . . . . . . . 456
13.3 Decorrelating Random Vectors and Multi-Dimensional Data . . . . . . . . . 467
13.4 Principal Components Analysis . . . . . . . . . . . . . . . . . . . . . . . . . 475
13.5 Chapter Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 484

Index 485
Taylor & Francis
Taylor & Francis Group
http://taylorandfrancis.com
Acknowledgments

I want to thank the many people who helped make this book possible.

My love and thanks go to my wife, Jill, for twenty-five years of love and friendship;
this book would not be possible without the wonderful family and home we have created
together. I love that we can enjoy quiet times at home or adventures together all over the
world. This book is better because of your careful editing.

To my children, Tucker, Charlotte, and Amelia: I treasure every moment with you,
whether it is trading funny memes, solving the New York Times crossword puzzle, battling
at Ping Pong, running our family 5Ks, trying to guess the winner of Survivor, reading to-
gether, or just working side-by-side on the couch.

I thank my parents, Larry and Agnes Shea, for their unwavering love and support. Mom,
you are missed every day.

I have been inspired by too many dedicated engineering educators to list them all. I
have to especially thank my Ph.D. advisor, Dr. Michael Pursley, for mentoring me on how
to conduct research, how to teach, and how to write technically. Thank you to Dr. John
Harris, who convinced me to develop the data-science course on which this book is based.
Thank you to Dr. Catia Silva, who was my co-instructor for one of the semesters teaching
the data-science course.

Several students provided feedback on the book, including Caleb Bowyer, Walter Acosta,
Cortland Bailey, Brennan Borchert, Alexander Braun, Patrick Craig, Justin Nagovskiy,
Allison Neil, Michael Russo, Dieter Steinhauser, Phillip Thompson, and Marisa Younger.

xi
Taylor & Francis
Taylor & Francis Group
http://taylorandfrancis.com
Preface

This book is an introduction to the foundations of data science, including loading and
manipulating data, data visualization, statistics, probability, and dimensionality reduction.
This book is targeted toward engineers and scientists, but it should be easily accessible
to anyone who knows basic calculus and the essentials of computer programming. By
leveraging this background knowledge, this book fits a unique niche in the books on data
science and statistics:
• This book applies a modern, computational approach to work with data.
In particular, it uses simulations (an approach called resampling) to answer statistical
questions.
– Many books on statistics (especially those for engineers) teach students how to
answer statistical questions using only analytical approaches that many learners
find difficult to understand. Most learners can easily understand how resampling
works, in contrast to some arcane formula.
• This book uses real data sets and addresses contemporary issues.
– Many statistics books use contrived examples that are small enough to print in
a book and work with using a calculator, but this results in data sets that are
unrealistic and uninteresting. The computational approach used in this book allows
the use of data sets from across the web to conduct statistical tests on topics from
the effects of socioeconomic factors on the spread of the COVID-19 virus to the
impact of state firearm laws on firearm mortality.
• This book provides a basic, but rigorous, introduction to probability and its
application to statistics.
– Some of the other books that use the resampling approach to statistics omit the
mathematical foundations because they are targeted toward a broader audience
who may not have the rigorous mathematical background of engineers and scien-
tists.

xiii
xiv Preface

• This book shows how to work with some of the most important tools in the
Python data-science stack, including:

– NumPy for working with vectors and matrices, as well as many types of numerical
functions,
– SciPy for working with random distributions and conducting statistical tests,
– Pandas for loading, manipulating, and summarizing data,
– Matplotlib for plotting data, and
– scikit-learn for accessing standard data sets and for advanced statistical process-
ing.

• This book was co-written with a book that covers linear algebra and its
application to data science using Python and NumPy.

– Techniques like dimensionality reduction require linear algebra. Although linear


algebra is not covered in this book, the companion book Introduction to Linear
Algebra for Data Science with Python provides the necessary background knowledge
with the same mix of analysis and Python implementation.
• This book provides a unique set of online, interactive tools to help students
learn the material, including:

– interactive self-assessment quizzes,


– interactive flashcards to aid in learning terminology,
– interactive Python widgets and animated plots.

Interactive elements are available on the book’s web site: fdsp.net.

View the quiz and flaschards for this preface at

fdsp.net/intro,
which can also be accessed using this QR code:

Credits: The image at the top of this page is made using the Wee People font made by
ProPublica: https://github.com/propublica/weepeople. Inspired by a Tweet by Matthew
Kay: https://twitter.com/mjskay/status/1519156106588790786.
1
Introduction

Welcome to Foundations of Data Science with Python! This chapter provides an introduction
to the book and its place in the field of data science. It then provides a brief introduction
to some of the tools that are used throughout the book. By the end of this book, you
will learn how to analyze and interpret data, formulate hypotheses about the data, perform
statistical tests, and communicate your findings accurately and effectively.

1.1 Who is this book for?


This book is targeted toward engineers and scientists, whether working or still in school.
Given this target audience, I assume that the reader has a basic working knowledge of:
• computer programming (knowing Python is helpful, but not required), and
• one-dimensional differential and integral calculus.
This book is written by an engineer with degrees in both electrical and computer engi-
neering. This book and its companion, Introduction to Linear Algebra for Data Science with
Python, were written to provide the main textbooks for a 4-credit, semester-long course for
engineers, taught in the Department of Electrical and Computer Engineering at the Uni-
versity of Florida. These books are intended to be a broad introduction to data science,
but they are also designed to replace courses in Engineering Statistics and Computational
Linear Algebra.

1.2 Why learn data science from this book?


This book uses a computational first approach to data science. You will learn how to
leverage the power of modern computers and scientific software to visualize, transform, and
simulate data. For instance, one of the main approaches used in this book is to conduct
statistical tests by carrying out simulations that draw samples from the data being analyzed.
This approach has the following benefits:
• We start working with real data sets quickly because this approach does not
require a lot of mathematical background and the computer does all the mathematical
manipulation and plotting.
• Simulation models are easy to create and understand. The results do not rely
on any arcane formulas but only the ability to build very simple simulations that draw
from the experimental data.

DOI: 10.1201/9781003324997-1 1
2 Introduction

• This approach is more general than the traditional approach. It does not rely
on the data coming from specific probability distributions, and the same simulation can
be used to generate frequentist or Bayesian statistics.

Interactive flashcards and self-assessment quizzes are provided throughout the book
to help learners master the material and check their understanding. The entire set of
interactive materials can be accessed on the book’s website at fdsp.net.
The interactive materials use spaced repetition to help readers retain knowledge as they
progress through the book. Starting with Chapter 2, the interactive chapter reviews also
give a random subset of review problems from earlier chapters. Research shows that spaced
repetition improves the retention of material.
Each chapter of the book ends with “take-aways” that help summarize the important
points from the chapter and address issues that may be topics of questions in data science
interviews.

1.3 What is data science?


Our world is filled with information. In fact, the amount of information we have access to
can often be overwhelming. We start by considering when information becomes data:

Definition
data
Collections of measurements, characteristics, or facts about a group.

Then a simple definition of data science is:

Definition
data science
The process of extracting meaning from data.

Data consists of data points:

Definition
data points
A collection of one or more pieces of information collected about a single
individual or entity.

Each data point may contain variables and features:


What is data science? 3

Definitions
variables
Particular characteristics, measurements, or facts that make up a data point.

features
Individual pieces of information in a data set. While variables typically
represent unprocessed or raw data, features can include both variables and
processed versions of the variables.

In the machine-learning (ML) literature, the term feature is often used for both raw
and processed data, especially if the data are used as the input for some ML process.
Until Chapter 12, we will primarily work with existing data sets and refer to the pieces of
information that make up the data points as variables. In Chapter 13, we will consider how
to transform data to create new features.
Variables and features may be either quantitative or qualitative:

Definitions
quantitative data
Numeric data. Quantitative data may be either discrete (such as the number
of people in a family) or continuous (such as grade point average).

qualitative data
Non-numeric data. Qualitative variables are generally non-numeric categories
that data may belong to (such as hair color). Some categories may have an
order associated with them, but the order does not imply a numeric nature
to the categories. For example, a survey question may have responses from
Strongly Disagree to Strongly Agree.

Examples of quantitative variables:

• height

• weight

• yearly income

• college GPA

• miles driven commuting to work


• temperature

• wind speed
• population

Examples of qualitative variables:

• hair color (blond, brown, black, red, gray, …)


4 Introduction

• current precipitation status (no precipitation, raining, snowing, sleeting, …)

• car type (sedan, coupe, SUV, minivan, …)

• categories of hurricanes or earthquakes


• model of smartphone used to access a mobile app

• blood type

As engineers and scientists, our goal is to make sense of the world and to use what we
learn to take action. Data science applies computational tools and mathematical methods
to process and transform data for the purpose of better understanding what the data can
(and cannot!) tell us about the world.
Data scientists often start with a research question:

Definition
research question
A question that can be answered using research, including data collection
and analysis.

For instance:

• Does more education translate to more wealth?


• Do state gun laws affect firearms mortality?

• Is the climate changing?

• How fast was the COVID-19 coronavirus spreading when it first became prevalent in
the United States in the Spring of 2020?

One of the goals of a data scientist is to take broad research questions and translate
them into questions that can be answered using data. One set of criteria (from Designing
Clinical Research by Hulley, Cummings, Browner, Grady, and Newman) has the acronym
FINER, which stands for Feasible, Impactful, Novel, Ethical, and Relevant.
In this book, I have tried to address these issues by working with existing data sets to
answer questions that are important and timely. For instance, here are some ways that the
questions above may be reformulated so that they can be answered with data:

• Instead of “Does more education translate to more wealth?”, a data scientist may ask a
more specific question, such as “For people in the United States, does post-baccalaureate
education increase median net family wealth?”.
• Instead of “Do gun laws affect firearm mortality?”, a data scientist may ask “Are state
permitless-carry laws associated with a difference in average firearm mortality rates?”.

• Instead of “Is the climate changing?”, a data scientist may ask “Has the average annual
temperature in Miami, FL increased over the past 40 years?”.

• In assessing the rate of spread of the COVID-19 virus, the data scientist may ask “Was
the number of cases growing exponentially in March of 2020? If so, what was the
exponential growth rate?”.
What data science topics does this book cover? 5

Terminology review

Interactive flashcards to review the terminology introduced in this section are


available at fdsp.net/1-3, which can also be accessed using this QR code:

1.4 What data science topics does this book cover?

FIGURE 1.1
Data science topics covered in this book.

Data science is a very broad topic. Data scientists use many different tools to make
sense of data, from databases to advanced ML algorithms. This book focuses on some of
the fundamental tools used to create meaning from data, and I have tried to introduce skills
and tools that will be useful to engineers and scientists in other contexts. The main topics
covered in this book are shown in Fig. 1.1 and described below:

• Simulation is used to emulate random phenomena and to carry out statistical tests.
• Visualization is used to transform data into graphical forms that help reveal trends in
the data or tell a story about the data.
• Data manipulation is the process by which data is loaded and prepared for analysis.

• Statistics is used to determine whether observations made from the data are meaningful
or could just be attributed to randomness in the data.

• Probability is used to create mathematical models for random phenomena; these mod-
els can be used to develop optimal estimators and make optimal decisions.
6 Introduction

• Data transformation consists of mathematical processing to achieve goals such as


creating new features or reducing the size of the data.

Throughout this book, I use two key tools to enable this computational approach:

• Jupyter is a web-based notebook environment that combines features of an integrated


development environment (IDE) with those of a word-processing or web-development
application. Jupyter notebooks can include text, mathematics, graphics, executable
program code, interactive widgets, and more. Specifically, this book uses JupyterLab.

• Python is a versatile programming language that has a rich set of libraries that support
data science activities.

1.5 What data science topics does this book not cover?
This book focuses on the foundations of data science, and there are many important topics
that could not be included. In particular, a prospective reader should know that:
• This is not a machine learning book. In particular, this book does not cover neural
networks at all.

• This book assumes that the reader knows the basics of linear algebra and how to work
with vectors and matrices in NumPy. However, I wrote this book simultaneously with
another book called Introduction to Linear Algebra for Data Science with Python that
provides the necessary background on linear algebra.
• This book does not cover many practical aspects of working with data, such as using
databases to retrieve and store data. Nor does it provide much coverage of different
approaches to “cleaning” data, such as dealing with missing or mislabeled data.

• This book does not provide comprehensive coverage of the libraries that are used, such
as NumPy, Pandas, and scikit-learn; rather, it focuses on showing how to use these
libraries for some foundational data science techniques.
• This book does not address many important issues related to the ethics of data science.

On the website for this book (fdsp.net), I provide a list of suggested “Next Steps” that
include books and online materials that address these important topics.

1.6 Extremely Brief Introduction to Jupyter and Python


The purpose of this and the following two sections is to briefly introduce users to Jupyter
and Python. The content here should be treated as an introduction to explore further and
is not meant to be comprehensive. There are a broad variety of tutorials on the web for
both of these topics, and links are provided for users who need additional instruction.
If you are already familiar with Jupyter and/or Python 3, feel free to skip ahead.
Extremely Brief Introduction to Jupyter and Python 7

1.6.1 Why Jupyter notebooks?


According to the Project Jupyter web page (https://jupyter.org), “The Jupyter Notebook
is an open-source web application that allows you to create and share documents that
contain live code, equations, visualizations and narrative text. Uses include: data cleaning
and transformation, numerical simulation, statistical modeling, data visualization, machine
learning, and much more”.
The reasons that Jupyter notebook was chosen for this book include:
• Jupyter notebooks can integrate text, mathematics, code, and visualization in a single
document, which is very helpful when conveying information about data. In fact, this
book was written in a series of over 140 Jupyter notebooks.
• Jupyter notebooks allow for an evolutionary approach to code development. Programs
can start as small blocks of code that can then be modified and evolved to create more
complex functions.
• Jupyter notebooks are commonly used in the data science field.

1.6.2 Why Python?


Python is a general-purpose programming language that was originally created by Guido
van Rossum and maintained and developed by the Python Software Foundation. Python
was chosen for this book for many reasons:
• Python is very easy to learn. Python has a simple syntax that is very similar to C,
which many engineers and scientists will be familiar with. It is also easy to transition
to Python from MATLAB scripting, which many engineers will be familiar with.
• Python is an interpreted language, which means that programmers can run the
code directly without having to go through extra steps of compiling their programs.
• Python interpreters are freely available and easy to install. In addition, Python
and Jupyter are available on all major operating systems, including Windows, MacOS,
and Linux.
• Python is popular for data science and machine learning. Python is widely
used for data science and machine learning in both industry and universities.
• Python has rich libraries for data science. Python has many powerful libraries
for data science and machine learning. In addition, Python has powerful libraries for
a broad array of tasks beyond the field of data science, which makes learning Python
have additional benefits.

1.6.3 How to get started with Jupyter and Python


Python and Jupyter are often packaged together in a software distribution, which is a
collection of related software packages. The creators of several Python software distributions
include additional Python software libraries for scientific computing. This book assumes
the use of the Anaconda distribution, which its creators bill as “The World’s Most Popular
Data Science Platform”1 .
Anaconda’s Individual Edition is freely available to download from the Anaconda website
at https://www.anaconda.com/products/individual. Choose the proper download based on
1 https://www.anaconda.com/, retrieved May 30, 2023.
8 Introduction

your computer’s operating system. You may also have to select a version of Python. This
book is based on Python 3, which means that any version of Python that starts with the
number 3 should work with the code included in this book. For instance, as of May 2023,
the Anaconda distribution included Python version 3.10.

WARNING

Python version 2 or Python versions after 3 may have syntax changes


that cause the programs in this book to not run without modification.

After downloading, install Anaconda however you usually install software(for instance,
by double-clicking on the downloaded file). Anaconda will install Python and many useful
modules for data science, as well as Jupyter notebook and JupyterLab.
Note:

The term “Jupyter notebook” refers to a file format (with .ipynb extension), while
“Jupyter Notebook” (with a capital N) refers to an application with a web interface
to work with those files. To help avoid confusion, I will write Jupyter notebook file
or simply notebook whenever referring to such a file, and we will use JupyterLab as
the web application for opening and working with such files.

As of January 2023, JupyterLab “is the next-generation web-based user interface for
Project Jupyter”. (from https://jupyterlab.readthedocs.io/en/stable/). The Jupyter Note-
book application offers a simple interface for working with notebooks and a limited number
of other file types. JupyterLab has a more sophisticated interface and can include many
different components, such as consoles, terminals, and various editors. The interface for
working with notebooks is similar in both, and most users will be able to use either one
interchangeably.

1.6.4 Getting organized


We are almost ready to start using Jupyter and Python. Before you do that, I recommend
you take a minute to think about how you will organize your files. Learning data science
requires actually working with data and performing analyses. This will result in you gen-
erating a lot of Jupyter notebook files, as well as a lot of data files. I suggest that you
create a folder for this data-science book (or for the course if you are using this as a course
textbook). This folder should be easily accessible from your home directory because that
is the location where JupyterLab will open by default. You may wish to add additional
structure underneath that folder. For instance, you may want to create one folder for each
chapter or each project. If you create separate folders for the data, I suggest you make
them subfolders of the one containing the notebooks that access that data.
An example layout is shown in Fig. 1.2.

1.6.5 Getting started in Jupyter


Let’s begin exploring JupyterLab using an existing notebook:
Extremely Brief Introduction to Jupyter and Python 9

FIGURE 1.2
Example directory structure for organizing files for working through the examples and
exercises in this book.

1) Download a Jupyter notebook file.

We will use the file “jupyter-intro.ipynb”, which is available on this book’s website at:

https://www.fdsp.net/notebooks/jupyter-intro.ipynb

If your browser displays the notebook as text, you will need to tell it to save it as a
file. You can usually do this by right-clicking or control-clicking in the browser window and
choosing to save the page as a file. For instance, in Safari 14, choose the “Save Page As…”
menu item. Be sure to name your file with a .ipynb ending.
Hint: If your file was saved to your default Downloads folder, be sure to move it to an
appropriate folder in your data-science folder to keep things organized!

2) Start JupyterLab.

JupyterLab can be started from the Anaconda-Navigator program that is installed with
the Anaconda distribution. Start Anaconda-Navigator, scroll to find JupyterLab, and then
click the Launch button under JupyterLab. JupyterLab should start up in your browser.
Alternative for command-line users: From the command prompt, you can start Jupyter-
Lab by typing jupyter lab (provided the Anaconda bin directory is on the command line
search path). Because setting this up is specialized to each operating system and command
shell, the details are omitted. However, details of how to set up the path for Anaconda can
be found at many sites online.
Your JupyterLab should open to a view that looks something like the one in Fig. 1.3.

WARNING

If you have used JupyterLab before, it may not look like this – it will
pick up where you left off!

The JupyterLab interface has many different parts:


1. The menu bar is across the very top of the JupyterLab app. I will introduce
the use of menus later in this lesson.
2. The left sidebar occupies the left side below the menu bar. It includes several
different tabs, which you can switch between by clicking the various icons on the
very far left of the left sidebar. In Fig. 1.3, the folder icon is highlighted, which
10 Introduction

FIGURE 1.3
The JupyterLab interface.

indicates that the file browser is selected. For this book, we will use the left
sidebar only to access the file browser.
3. The main work area is to the right of the left sidebar. The main work area will
usually show whatever document you are working on. However, if you have not
opened any document yet, it will show you different types of notebooks that you
can open and other tools that you can access. To start a completely new Jupyter
notebook file that can run Python 3 code, you could click on the Python 3 icon
under Notebooks. For now, you do not need to do that.
Detailed documentation for JupyterLab is available at

https://jupyterlab.readthedocs.io/.

3) Navigate to the downloaded notebook.

Use the file browser in the left sidebar of JupyterLab to navigate to the downloaded file.
If the file browser is not already showing your files, click on the folder icon (on the very
left-hand side of the window) to switch to it.
Navigation using the file browser should be similar to navigating in most file selection
boxes:

• Single click on items to select them.

• Double click on a folder to navigate into it.

• Double click on a file to open it.


Random documents with unrelated
content Scribd suggests to you:
MAJOR-GENERAL M’CLERNAND.
Major-General John A. McClernand was a lawyer by
profession, and had figured prominently as a leading Democratic
politician from Illinois. He was a leader of the Douglas Democrats,
and did battle for them valiantly at Charleston. At the outbreak of the
war he took sides manfully for the Union, and shortly afterward was
nominated a Brigadier-General of Volunteers. In the Belmont fight
he gave evidence that he possessed good military capacity, and
during his administration of military affairs at Cairo he secured the
good will of the men under his command. In the reconnoissance in
the rear of Columbus, during the advance upon Fort Henry, and at
the grand battle before Fort Donelson, General McClernand
manifested superior military ability. For his gallantry on these
occasions he was, on the 21st of March, 1862, made by Congress a
Major-General of Volunteers, and accompanied the advance up the
Tennessee river toward Savannah. At the battle of Pittsburgh
Landing he was highly distinguished.

After spending two days devoted to the care of the wounded, and
the burial of the dead, the fort was blown up and completely
destroyed, the rifle pits levelled, and a hundred wagons which had
been captured, were burnt. On the 18th, General McClernand
embarked with the main body of his troops, and proceeded down the
Arkansas river to Napoleon, where a conference was held with
General Grant and Admiral Porter, and future operations were
planned.
Meantime an expedition of light-draught steamers, under
Lieutenant-Commander J. G. Walker, and a body of troops led by
General Gorman, had proceeded up the White river, and captured
the towns of Des Arc and Duval’s Bluff.
General McArthur’s corps of General Grant’s army, left Memphis
on the 20th of January on transports, and landed at Young’s Point,
on the west side of the river, about nine miles above Vicksburg. Here
the greater part of the fleet was concentrated; and on the 2d of
February, General Grant arrived, and assumed command of the
army. At this point a canal had been commenced by General
Williams, previous to the unsuccessful attack on Vicksburg the year
before, for the purpose of effecting a passage for vessels across the
peninsula in front of Vicksburg out of range of the enemy’s guns.
The attack on Vicksburg, from up the river, had demonstrated the
strength of its defensive works on the north, and convinced General
Grant that they were too strong to be carried without a very heavy
loss. The first step for him to accomplish, therefore, was the
transportation of his army below the city, in order to make an attack
from the south. The passage by the river was too hazardous to be
attempted. The formidable batteries on the river front at Vicksburg
were capable of destroying all the transports. Work was therefore
recommenced on the canal. While this work was in progress, the
river continued to rise rapidly, and great labor was required to keep
the water out of the canal, and also out of the camps of the laborers
and soldiers. In addition, the rain was incessant, and the magnitude
of the work was, from these causes, greatly increased. The earth
taken out of the excavation was placed on the west side, and thus
formed an embankment or levee, which it was supposed would
prevent the water from flooding the country.
While a portion of General Grant’s forces were employed in cutting
the canal at Young’s Point, their commander, firmly intent on
accomplishing the great enterprise before him, was industriously
employed in the prosecution of other plans, which might be
consummated in the event of a failure in the canal at the peninsula. A
channel was cut from the Mississippi into Lake Providence, on the
west side of the Mississippi, and another into the Cold Water river on
the eastern bank, by way of the Yazoo Pass.
While these operations were in progress, a daring enterprise was
undertaken by Colonel Charles E. Ellet, commander of the ram
steamer Queen of the West, by which he hoped to destroy a
formidable rebel steamer called the City of Vicksburg, then lying
under the guns of the fortifications at that place.
The Queen of the West had been previously provided with all the
arrangements deemed necessary to insure the complete success of
the dangerous undertaking. Three hundred bales of cotton had been
procured further up the river and placed on board, particularly about
the machinery, in order to save her from any serious injury by shot
and shell from the rebel batteries. Rear-Admiral Porter had given
orders that she should proceed down to Vicksburg, destroy the rebel
steamboat City of Vicksburg, lying opposite the city, and then run
past the lower rebel batteries. The Colonel was directed to keep close
to the right bank going down, to have all his lights on board
extinguished—as it was intended that she should run the gauntlet in
the darkness—and having safely passed the batteries, to anchor
below the mouth of the canal and there wait for further orders.
The Colonel started with the ram from above the bend at half past
four o’clock in the morning. It was about six o’clock, just as the sun
was rising, when the ram rounded the point of land lying opposite
Vicksburg. She had only men enough on board to work her, it having
been arranged that the remainder of the crew would cross the point
of land and get on board of her below after she had passed the
batteries. When rounding the point she was distinctly seen by the
rebels. They immediately opened a heavy fire from several of their
batteries, which crowned the crests of the bluffs about the city. The
Queen slowly and steadily proceeded down the river under a heavy
fire from those batteries, until she reached a point opposite the spot
where the steamboat City of Vicksburg was lying. Colonel Ellet saw
that the steamboat was lying in almost the same position as was the
rebel ram Arkansas when he ran into her with the Queen of the West.
If the rebel steamboat should be struck as the ram was running down
the river, the prow, instead of penetrating her, would be inclined to
glance, and the full force of the blow would thus be lost. Wishing to
make the shock as effective as possible, when the ram had reached
the proper position the Colonel turned her partly around, so as to
face the city, and then made across the river straight for the fated
steamboat. The rebels, who had crowded on the banks, scampered
off in the most affrighted manner from the shore, and sought safety
in the city. The ram still went steadily on to the execution of her
destructive errand. She struck the rebel steamboat forward of the
wheel-house; but at the moment of collision the current caught the
stern of the ram and swung her round so rapidly that nearly all the
momentum of the blow was lost. To set the rebel steamboat on fire
was part of the arrangement. That portion of the programme was
intrusted to Sergeant J. H. Campbell. He was directed to fire the
forward guns, loaded with combustible balls saturated with
turpentine. As the ram swung round he was ordered to fire them.
Just at that moment a 64-pound shot from one of the rebel batteries
came crashing into the barricade of cotton near him; but the brave
Sergeant did not hesitate a moment in the execution of the order.
The guns were fired, a tremendous blaze was vomited forth from
them, and the rebel steamboat was in flames.
About the same time the ram was found to be on fire. A shell from
shore had set her on fire near the starboard wheel, while the
discharge of the guns with the combustible balls had fired the cotton
on her bow. Both steamboats were thus ablaze at the same time. The
flames spread rapidly on both vessels. The smoke from the front of
the ram rushed into her engine-room and threatened to suffocate the
engineers. Those on board the rebel steamboat did all they could do
to extinguish the flames on their boat. This they soon accomplished.
Colonel Ellet had intended to strike the rebel steamboat in the stern,
and thus finish the work of demolition; but the spreading flames on
the Queen of the West made it necessary for him to attend to the
safety of his own vessel. He therefore ran down stream, and set all
hands on board at work extinguishing the flames. Though the cotton
had been wet before starting, the fire was extending rapidly, and
several burning bales were thrown overboard in order to save the
ram. She then anchored below the mouth of the canal, where she
awaited further orders.
All this time, both when approaching the city and leaving it, the
rebel batteries were blazing away at the Queen of the West with light
and heavy guns. It was a very exciting scene. About one hundred and
twenty shots were fired from the batteries; but the ram was struck
only twelve times, and sustained no material injury.
The Queen of the West now proceeded down the Mississippi, and
when below Natchez, burned three small rebel steamers, the Moro,
Berwick Bay, and A. W. Baker, laden with stores for the army at
Vicksburg. After cruising for two weeks in the Atchafalaya, the Red
river, and other tributaries of the Mississippi, inflicting serious injury
on the rebel commerce, and capturing several vessels, she finally
ventured up the Black river, and captured the rebel steamer Era. She
proceeded to Fort Taylor, some fifty miles from the mouth of the
river, where she grounded on a bar, exposed to the fire of the guns of
the fort, and her crew was compelled to abandon the vessel, which
fell into the hands of the enemy. Captain Ellet and most of the men
succeeded in reaching the Era, and effected their escape.

When the work on the canal through the peninsula had


approached its completion, and the huge iron scoop of the dredging
machine had commenced demolishing the barrier which intervened
between the bed of the canal and the “Father of Waters,” an
unforeseen occurrence, which could not be guarded against, crushed
the enterprise.
Owing to heavy rains and the rapid rise of the Mississippi above
and opposite Vicksburg, the head of the canal gave way, and the
water poured in at a tremendous rate. The force of the current,
however, did not break the dam near the mouth of the canal, but
caused a crevasse on the western side, through which the water
flowed in such profusion as to inundate the lower part of the
peninsula to the depth of four or five feet. When the fracture
occurred a number of soldiers were on the levee, and were thrown
into the torrent, but no lives were lost. All attempts to repair the
mischief proved ineffectual, and the troops were removed to
Milliken’s Bend, fifteen miles above.

On the 27th of February, Admiral Porter dispatched what was


called a dummy Monitor, to run the Vicksburg batteries, in order to
ascertain their exact location. This contrivance was an old flatboat,
with flour-barrels for smoke stacks, and a couple of large hogsheads
to represent Monitor turrets. It ran the fortifications in gallant style,
and drew the fire of the rebel guns, without creating a suspicion of
the true character of the vessel. The rebel authorities, fearful of the
capture of the Indianola, then in an exposed position undergoing
repairs, caused that vessel to be blown up to prevent her from falling
into the hands of the Federals, and thus uselessly sacrificed the finest
iron-clad they had on the western waters.
The prudent forethought of General Grant exhibited by his
employing a portion of his men in cutting channels from the
Mississippi to Providence lake on the west side, and to Moon lake on
the east side, was now made apparent, and those works were
progressing rapidly.
Lake Providence is a few miles south of the boundary line between
Arkansas and Louisiana. It is situated in Carroll parish, Louisiana,
about one mile west of the Mississippi river, and about seventy-five
miles above Vicksburg. It is about six miles in length. Two streams
flow out of the lake to the south, Moon bayou and Tensas river. The
former, after running about a hundred miles, unites with the latter.
The two continue south, and unite with the Washita, and are called
after the junction, Black river. By cutting a channel from the
Mississippi to Lake Providence, General Grant thought a
communication might be had through that lake down the Tensas and
Black into the Red river, and thence through the Atchafalaya, with
General Banks at New Orleans. This route avoided the batteries of
Vicksburg and Port Hudson. The canal to the lake was finished so as
to let in water on the 16th of March. The flood was so great as to
inundate a large district of country, some of which was fine land for
growing cotton. Some boats passed into Lake Providence, but the
uncertainty of the channel of the Tensas river, and the interest which
was now excited by the Yazoo Pass expedition, together with the
unimportant results to be anticipated by removing a large force to
the Red river or below, caused a diversion from this route to others
presenting more certain prospects of success against Vicksburg.
Eight miles below Helena, in Arkansas, and on the opposite side of
the river, is a little lake, known as Moon lake. The passage from the
Mississippi across the lake to the mouth of the Yazoo Pass is about
eight miles; thence through the Pass proper to the Coldwater river,
twelve miles. The Coldwater, a narrow stream, runs south, empties
into the Tallahatchie, which continues to flow south, and unites with
the Yallobusha, forming the Yazoo river, which empties into the
Mississippi a few miles above Vicksburg.
Another important operation took place on the 14th of March,
which had much to do with the success of General Grant’s
movements. Admiral Farragut, with his fleet, attacked Port Hudson,
and the flagship succeeded in running past the batteries and arriving
before Warrenton, when he communicated with the fleet above.
Shortly after this Admiral Porter succeeded in running some of his
fleet down to the assistance of Farragut, and the united fleets began
operating upon the river between Vicksburg and Port Hudson,
cutting off the communications of the rebels with Louisiana, and
making important military movements on the Louisiana shore.
An attempt to pass the rebel batteries at Vicksburg, was made by
the Union rams Lancaster and Switzerland, on the 25th of March,
without success. As soon as they came within range, the rebels
opened a tremendous fire. The Lancaster was struck thirty times.
Her entire bow was shot away, causing her to sink immediately. All
the crew except two escaped. The Switzerland was disabled by a 64-
pound ball penetrating the steam-drum. She floated down, the
batteries still firing, and striking her repeatedly, until finally the
Albatross ran alongside and towed her to the lower mouth of the
canal.
An expedition proceeded down the Coldwater, on the 2nd of April,
consisting of a portion of General Sherman’s and General
McClernand’s corps, under General L. F. Ross, with eighteen
transports and five small gunboats, and arrived at the mouth of the
river without obstruction. They proceeded down the Tallahatchie, to
its junction with the Yallobusha, which there forms the Yazoo, near
which point is the village of Greenwood. On a peninsula near by, the
rebels had erected a fortification. It consisted of a single line of
breastworks facing westerly, composed of cotton bales and earth, and
flanked on the right by a battery of three heavy guns fronting the
river. Other field-pieces were in position on the works. On the right
flank of the line, a defence or raft of logs had been constructed, to
serve as a blockade of the river. Directly in front of the breastworks
was a deep slough, extending across the peninsula, and admirably
serving the purpose of a ditch. The slough was close to the base of the
works at the upper end, but gradually receded from them at the
lower, where it was several hundred yards distant. Beyond the slough
there was an almost impenetrable canebrake, backed by an extensive
forest.
The reduction of this fort was an inevitable necessity, before the
expedition could proceed further, and the gunboat Chillicothe,
Lieutenant Foster, was sent forward on the morning of the 11th of
April to reconnoitre. The vessel approached the fortification, and
fired several shots, but was soon struck four times by heavy rifle
shots.
At the same time detachments from the Forty-sixth and Forty-
seventh Indiana regiments were sent out to feel the Confederate
position on the land side. A considerable body of the enemy’s
skirmishers were encountered, who were driven across the slough
and into the works, when the detachments were withdrawn. In the
afternoon the Chillicothe was ordered to engage the fortification.
After she had fired seven rounds, a 64-pound shell from the enemy
passed through a half-open port striking upon the muzzle of a gun, in
which a shell had been placed preparatory to cutting the fuse. Both
shells exploded at once, by which three men were killed and eleven
wounded. At this time orders were received to withdraw from the
engagement. During the ensuing night a force was sent to throw up a
battery facing the enemy’s works, west of the slough, and in the edge
of the timber. A single 30-pound Parrott gun was mounted, and the
work concealed by brush from the view of the enemy. Subsequently
another gun was mounted. No attack was made on the 12th, in
consequence of the absence of the mortar boats. After some delay, on
the 13th, the engagement was commenced about half past ten A. M. by
the land batteries. The gunboats Chillicothe and DeKalb soon after
approached and opened their fire. It now appeared that the
fortification mounted a rifled 64 Parrott, and three 24-Dahlgrens,
and a small field battery. These guns were protected by a parapet
composed of seven tiers of cotton bales, covered on the outside with
eight feet of earth. The contest was bravely maintained for some
time, when the fire of the enemy was suspended, but no disposition
to surrender was shown. The gunboats and battery kept up the fire,
but without any success in reducing the works. The Chillicothe was
struck thirty-four times, but not severely injured. The DeKalb
suffered more, in consequence of some shot penetrating her
casemates, by which one man was killed and five wounded.
The impracticable nature of the land approaches rendered any
attempt on the part of the military futile, and the expedition was
compelled to retire.
An expedition under Admiral Porter, consisting of the heavier
gunboats of his flotilla was undertaken about this time accompanied
by transports, for the purpose of reaching the Yazoo river below Fort
Pemberton, and Greenwood, and above Haines’ Bluff. The route of
this expedition was up the Yazoo to Cypress bayou, thence into
Steele’s bayou, and through Cypress lake to Little Black Fork and
Deer creek. These waters were found to be impenetrable to the
Federal vessels, and that expedition also proved a failure.
PASSING THE VICKSBURG BATTERIES.
April 16–22, 1863.

General Grant having now fully determined to transport his army


below Vicksburg, the cooperation of the naval commander was
obtained: and on the 16th of April, three transports, under cover of
the gunboats, were in readiness to descend the river. At half-past ten
at night Admiral Porter’s vessels started on their perilous expedition
down the river.
The vessels comprising the expedition started in the following
order, fifty yards apart: Benton, Lieutenant-Commander Green;
Lafayette, Captain Walker, with the General Price lashed on the
starboard side; Louisville, Lieutenant-Commander Owen; Mound
City, Lieutenant Wilson; Pittsburg, Lieutenant Hall; Carondolet,
Lieutenant Murphy, and Tuscumbia, Lieutenant Commander Shirk,
with the tug Day, which was lashed to the Benton. The three army
transports were in the rear of these vessels, and the Tuscumbia was
placed astern of all.
Two of the transports, when the firing became heavy, attempted to
run up stream; but Lieutenant Commander Shirk drove them back,
and stayed behind them until the Forest Queen was disabled. He
then took her in tow, and placed her out of reach of the enemy’s shot.
All the vessels except the Benton took coal barges in tow, and all
except the Lafayette brought them safely past the batteries. Having
the Price alongside the Lafayette did not manage very well, and the
coal barge got adrift, but was picked up at New Carthage. The
Louisville, Lieutenant Commander Owen, lost hers in the melee, but
picked it up again. The Benton fired over eighty shell, well directed,
into the town and batteries.
The Pittsburg, Mound City and Tuscumbia were more fortunate
than the others in not turning round as they came by, although no ill
results happened to those vessels which did turn. The pilots were
deceived by a large fire started on the side opposite to Vicksburg by
the rebels, for the purpose of showing the vessel more plainly. Fires
being started on both sides of the river at once, the vessels had some
narrow escapes, but were saved in most instances, by the precautions
taken to protect them. They were covered with heavy logs and bales
of wet hay, which was found to be an excellent defence.
No one on board of the transports was injured; and, encouraged by
the success of this enterprise, General Grant ordered six more to be
prepared in like manner for running the batteries. On the night of
the 22d of April the Tigress, Anglo Saxon, Cheeseman, Empire City,
Horizona and Moderator, left Milliken’s Bend, and all passed in
safety but the Tigress, which received a shot in her hull below water-
line, and sunk on the Louisiana shore.
CAPTURE OF GRAND GULF, MISS.
April 29-May 3, 1863.

On Wednesday, the 29th of April, a part of the gunboat fleet under


Admiral Porter, consisting of the Benton (flagship), Lafayette,
Mound City, Pittsburg, Carondolet, Tuscumbia and Louisville, left
Hard Times, and steaming down below Coffee Point, engaged the
rebel batteries at Grand Gulf, just at the confluence of the Big Black
and the Mississippi. The engagement commenced at eight o’clock
and lasted until half-past two.
The enemy had four batteries at Grand Gulf, one on the rock
around which the waters of the Big Black flow into the Mississippi,
and three below, about midway between the water and the summit of
the bluffs. In the former they had placed four heavy guns, and in the
three latter two and three each, with parapets, embrasures and rifle
pits. The upper guns were very large, throwing shot and shell
weighing one hundred pounds. The lower guns were mostly thirty-
two-pounders.
The Benton opened the fight, followed by the other gunboats in
rapid succession. At first they stood off at long range, and fired at a
distance of a mile or more; but as the fire became warm they pressed
closely to the bluff, and passed and repassed the batteries, sending
broadsides upon the fort whenever they came in position. For six
hours where the gunboats firing and receiving fire, until the
gradually ceasing explosions of the enemy indicated their desire to
terminate the engagement.
The Benton suffered considerably in her upper works, not less
than a half dozen shots passing entirely through her. One shell
exploded in her porthole, killing five men. The Tuscumbia was
disabled. Other gunboats were more or less injured, but not
seriously.
On the same day Admiral Porter sent the following report to the
Secretary of the Navy:

Hon. Gideon Welles, Secretary of the Navy:—


I have the honor to inform you that, by an arrangement with General Grant, I
attacked the batteries at Grand Gulf this morning, which were very formidable.
After a fight of five hours and thirty minutes we silenced the lower batteries, but
failed to silence the upper one, which was high, strongly built, and had guns of very
heavy calibre. The vessels were unmanageable in the heavy current. It fired but
feebly toward the last, and the vessels all laid by and enfiladed it, while I went up a
short distance to communicate with General Grant, who concluded to land the
troops and march over to a point two miles below Grand Gulf. I sent the Lafayette
back to engage the upper batteries, which she did, and drove the soldiers out of it,
as it did not respond after a few fires. At six p. m. we attacked the batteries again,
and under cover of the fire all the transports passed by in good condition. The
Benton, Tuscumbia, and Pittsburg were much cut up, having twenty-four killed
and fifty-six wounded, but they are all ready for service.
We land the army in the morning on the other side to march on Vicksburg.
DAVID D. PORTER, Acting Rear Admiral.

On the 30th of April General Grant, with the three selected corps
de armeé, viz.:—the Thirteenth, General McClernand; the Fifteenth,
General Sherman, and the Seventeenth, General McPherson, crossed
from the Louisiana side of the Mississippi river and landed at
Boulinsburg.
The total number of killed in the fleet was twenty-six, and the
wounded fifty-four.
The bombardment was terrific, the gunboat men exhibiting a
coolness, courage and determination which it seemed nothing could
resist. The rebels stood bravely to their guns, but the steady and
heavy fire of the iron-clads drove them again and again. All the boats
were struck repeatedly; but the Tuscumbia was the only one
materially damaged. She had her hogchains cut away, and was
otherwise so badly damaged that it was deemed advisable to remove
her from the scene of action.
Finding it useless to protract the contest when the object to be
attained could be reached by another way, the gunboats moved out of
range and prepared to run the blockade at night. The usual
precautions were taken to prevent casualties, and each gunboat
carried all the troops it could conveniently accommodate. The
transports were filled with troops, and all the barges crowded—so
eager were the soldiers to take part in the exciting scene. The
gunboats started first, and were subjected to a severe fire, for it was
almost as bright as midday; the moon shining from a cloudless sky.
When the transports appeared the concentrated fire of the rebel
batteries was directed to them; but none of the vessels were disabled.
The Cheeseman had six or seven horses killed by the explosion of a
shell; but no lives were lost, as far as heard from.
General Grant was aboard a tug during the fight, and directed the
movement of troops, under cover of the gunboat fire. The forces
landed at Bayou Pierre.
BOMBARDMENT OF HAINES’ BLUFF.

Simultaneous with the land attack on Grand Gulf, General


Sherman made a demonstration on Haines’ Bluff on Wednesday
morning, April 29. A considerable force was embarked on the
transports, and preceded by the iron-clad Choctaw and all the
wooden gunboats in the Yazoo, proceeded up that stream. The
bombardment at the Bluff was chiefly carried on by the gunboats
Choctaw, DeKalb, Black Hawk, Romeo, Linden and three mortar
boats. The former was struck fifty-two times, and had her upper
works pretty badly battered. Her pilot house was struck by an eight-
inch solid shot, which penetrated the extreme top, but fortunately
injured none of the occupants. Her turret was struck repeatedly, but
the shot all glanced off. She was also penetrated by three shots below
the water line, one shot entering three feet below the surface of the
water. Another shot penetrated her casemates and floundered on her
deck. It was supposed to be a shell, and all hands beat a hasty retreat,
except Chief Engineer Baldwin, who ran up, seized it and threw it
overboard. General Sherman landed his forces on the south bank of
the Yazoo. The main object of the expedition was to prevent the
enemy from sending reinforcements to Port Gibson. The rebels
displayed a large force, and anticipated a battle. The expedition
returned on the 7th of May.

On the third of May, Admiral Porter took possession of the forts at


Grand Gulf. The details of the occupation are narrated in his report
to Secretary Welles, of the same date:

Flagship Benton, Grand Gulf, Miss., May 8, 1863.

Hon. Gideon Welles, Secretary of the Navy:


Sir—I have the honor to report that I got under way this morning with the
Lafayette, Carondolet, Mound City and Pittsburg, and proceeded up to the forts at
Grand Gulf, for the purpose of attacking them if they had not been abandoned. The
enemy had left before we got up, blowing up their ammunition, spiking their large
guns, and burying or taking away the lighter ones. The armament consisted of
thirteen guns in all. The works are of the most extensive kind, and would seem to
defy the efforts of a much heavier fleet than the one which silenced them. The forts
were literally torn to pieces by the accuracy of our fire. Colonel Wade, the
commandant of the batteries, was killed; also his chief of staff. Eleven men were
killed that we know of, and our informant says that many were wounded, and that
no one was permitted to go inside the forts after the action except those belonging
there.
We had a hard fight for these forts, and it is with great pleasure that I report that
the navy holds the door of Vicksburg. Grand Gulf is the strongest place on the
Mississippi. Had the enemy succeeded in finishing the fortifications no fleet could
have taken them.
I have been all over the works and found them as follows:—One fort on a point of
rocks seventy-five feet high, calculated for six or seven guns, mounting two seven
inch rifles, and one eight-inch and one Parrott gun on wheels, which was carried
off. On the left of this work is a triangular work, calculated to mount one heavy
gun. These works are connected with another fort by a covered way and double
rifle pits extending one quarter of a mile, constructed with much labor, and
showing great skill on the part of the constructor. The third fort commands the
river in all directions. It mounted one splendid Blakely one hundred-pounder, one
eight-inch and two thirty-pounders. The latter were lying bursted or broken on the
ground.
The gunboats had so covered up everything with earth that it was impossible to
see at first what was there, with the exception of the guns that were dismounted or
broken.
Every gun that fell into our hands was in good condition, and we found a large
quantity of ammunition.
These are by far the most extensively built works, with the exception of those at
Vicksburg, I have seen yet, and I am happy to say that we hold them.
I am dismounting the guns, and getting on board the ammunition.
Since making the above examination new forts have been passed nearly finished.
They had no guns, but were complete as regards position, and had heavy field
pieces in them.
DAVID D. PORTER, Acting Rear Admiral,
Commanding Mississippi Squadron.
CAPTURE OF PORT GIBSON, MISS.
May 1, 1863.

Two days after the bombardment of Grand Gulf by Admiral


Porter’s fleet, General Grant’s forces made a successful attack on
Port Gibson, a point six miles in the rear of Grand Gulf, which
compelled the rebels to evacuate the latter place. General Grant sent
the following dispatch to General Halleck, dated May 3:

Grand Gulf, Miss., May 3, 1863.

Major-General Halleck, General-in-Chief:

We landed at Boulingsburg April 30, moved immediately on Port Gibson, met


the enemy eleven thousand strong, four miles north of Port Gibson, at two o’clock
A. M. on the 1st inst., and engaged him all day, entirely routing him, with the loss of
many killed, and about five hundred prisoners, beside the wounded. Our loss is
about one hundred killed, and five hundred wounded.
The enemy retreated towards Vicksburg, destroying the bridges over the two
forks of the Bayou Pierre. These were rebuilt, and the pursuit has continued until
the present time.
Besides the heavy artillery at this place, four field pieces were captured and some
stores, and the enemy were driven to destroy many more.
The country is the most broken and difficult to operate in I ever saw.
Our victory has been most complete, and the enemy are thoroughly demoralized.
Very respectfully, U. S. GRANT,
Major-General Commanding.

Governor Yates, of Illinois, writing from Grand Gulf, on the


following day, gives a glowing account of the operations of the
Federal army:
“Our arms are gloriously triumphant. We have succeeded in winning a victory
which, in its results, must be the most important of the war. The battle of May 1
lasted from eight o’clock in the morning until night, during all of which time the
enemy were driven back on the right, left and centre. All day yesterday our army
was in pursuit of the rebels, they giving us battle at almost every defensible point,
and fighting with desperate valor. Last night a large force of the enemy was driven
across Black river and General McClernand was driving another large force in the
direction of Willow Springs. About two o’clock yesterday I left General Logan, with
his division in pursuit of the enemy, to join General Grant at Grand Gulf, which the
enemy had evacuated in the morning, first blowing up their magazines, spiking
their cannon, destroying tents, etc. On my way to Grand Gulf I saw guns scattered
all along the road, which the enemy had left in their retreat. The rebels were
scattered through the woods in every direction. This army of the rebels was
considered, as I now learn, invincible; but it quailed before the irresistible assaults
of Northwestern valor.”
GRIERSON’S RAID.
April 17–May 2, 1863.

For a long time Colonel Grierson’s ambition had been to lead the
cavalry force under his charge into the enemy’s country. At last he
received an intimation from General Grant’s headquarters that his
desire would be gratified. Colonel Grierson commanded the First
brigade of cavalry under General Grant. This force had been for some
time occupying Lagrange, Tenn., which is a small town on the
Memphis and Charleston railroad, about fifty miles east of the city of
Memphis, and four miles west of the junction of the Mississippi and
Charleston railroads. When Colonel Grierson first received
permission to move with his force into Mississippi, one of the chief
objects of the expedition was to cut off the means of communication
between the rebel army of the West, and that of General Bragg, then
in Middle Tennessee; but when the expedition was once begun, it
branched off into many unthought-of directions, and ended in being
one of the most brilliant, as well as important feats of the war.
On the morning of the 17th of April, Colonel Grierson received
orders from General Grant to move his force out on the Ripley road;
accordingly, his brigade, consisting of the Sixth Illinois cavalry,
commanded by Lieutenant-Colonel Loomis; the Seventh Illinois
cavalry, Colonel Edward Prince; and the Second Iowa cavalry,
Colonel Edward Hatch, obeying the directions they had received,
bivouacked for the night on a plantation a few miles northwest of the
town of Ripley. During the night five guerrillas were captured by the
Union men. On the morning of the next day the march began; the
main body of Colonel Grierson’s men proceeded in a southerly
direction, while one regiment, the Second Iowa, crossed the
Tallahatchie, and went in a southeasterly direction. On both sides of
the river the enemy’s pickets were posted in all directions,
endeavoring to prevent the Union soldiers from crossing, and there
was constant skirmishing between them and the rebels. The pickets
were constantly driven in; and an attempt to fire the bridge at New
Albany was prevented by the rapid movements of Grierson’s men. At
the close of the day the Union troops had accomplished their
contemplated march, and were stationed as follows: the Sixth and
Seventh Illinois regiments were encamped on a plantation a few
miles south of New Albany, and the Second Iowa about four miles
east of the same place. The Second Iowa, during the night, repulsed a
severe attack of the enemy. On the morning of the 19th, Colonel
Grierson dispersed his troops in various directions, with a view to
mislead the enemy, and cause him to suppose that the main object of
the expedition was to break up the various military organizations in
that part of the country. Accordingly, one detachment marched to
the eastward, another moved back toward New Albany, and a third
marched northwest towards King’s Bridge; and the enemy was thus
completely puzzled and in total ignorance of the real destination of
the Union forces.
Colonel Grierson himself, with the main body of his command,
marched in a southerly direction, and were subsequently joined by
the remainder of the force, when they took the road to Pontotoc.
There they met a detachment of the rebels, who fled before them,
after the exchange of a few shots, and were hotly pursued, and driven
through the town. Their entire camp equipage was captured, and a
large store of salt, which was destroyed. The march was then
continued till about eight o’clock at night, and the men encamped at
a point on the road leading to Houston, a few miles south of the
Pontotoc. At an early hour on the following morning the reveille was
sounded. Major Lull of the Second Iowa, with about one hundred
and fifty picked men, and one piece of artillery, was then sent back to
Lagrange in charge of all the prisoners and captured property which
had been taken from the rebels, in order that the force might be
relieved of all incumbrance, and the enemy made to suppose that
Colonel Grierson was retracing his steps.
At five o’clock on the morning of the 21st, Colonel Hatch was
ordered with his command to proceed up the Columbus road, and
destroy as much of the Mobile and Ohio railroad as was possible; and
to attack Columbus These orders were successfully carried out; and
Colonel Hatch, with the troops under his command returned to
Lagrange, and thus aided in still further deceiving General Chalmers
(who was in command of the rebels at this point) in regard to the
movements of Colonel Grierson.
In the mean time the remainder of the Union forces had continued
their march, and reached Starkville, where they captured and
destroyed a rebel mail which had arrived; and set fire to and utterly
destroyed one of the finest tanneries in the country, which they
reached after continuing their march for five miles in a southerly
direction.
On the following day, the 22d, the march was not only
disagreeable, as the men were often compelled to swim their horses
through streams and lead them over blind marshes, but extremely
perilous; for often horse and rider would sink into the marsh
together, and though the men escaped with life, the hapless animals
often disappeared and were lost. With unparalleled fortitude the men
pushed on; and at ten o’clock the next morning they reached
Philadelphia, Miss. At this place the mail was taken from the post-
office, and destroyed, but nothing else was injured in any way.
On the following day the march was vigorously prosecuted. A
battalion was sent by the Southern railroad to Decatur and Newton,
where they were joined the same night by the main body, under
Colonel Grierson. Two trains of cars were captured at Newton, laden
with every description of commissary stores, and a large quantity of
ammunition and loaded shell. All were destroyed, and the
locomotives rendered unfit for any further service. The march was
then resumed, and continued till the 25th, when a halt was made at a
plantation a few miles west of Montrose, the men having fired every
bridge which they passed on the way. From this place the route was
slightly changed, and the cavalry pressed on in a more southerly
direction.
At Raleigh they halted for the night, and a scout who had been sent
out to cut the telegraph wires on the Southern railroad between Lake
Station and Jackson, was met by the enemy, and questioned as to the
whereabouts of Grierson’s men. The rebels were then on the direct
road to the camp, and not more than fourteen miles distant; but the
scout, with admirable self-possession, parried their questions, and
succeeded in misleading them as to the position of the Union troops,
and then escaped and hastened back to camp in time to give
information of the enemy’s proximity. Colonel Grierson immediately
moved his men across Leaf river, and destroyed the bridge to prevent
the rebels following and attacking him in the rear. They then
marched on to Westville, and swam their horses across Reaul river,
at a point ten miles from Westville. The advance, under Colonel
Prince, had by this time reached Hazlehurst station, where they
captured a train of forty cars,—four of which were filled with shell
and ammunition, and the remainder with commissary stores. As the
march continued, the cavalry came upon a team carrying a 32-pound
Parrot gun, which was then on its way to Fort Gibson. The piece was
captured and spiked. Two detachments from the main body had been
doing serious damage to the rebels, burning cars, water tanks, and a
great deal of other property.
At early dawn on the 28th, the advance moved upon Brookhaven,
and entered the town so suddenly that two hundred rebels were
taken prisoners, before they had recovered from their surprise at
finding themselves confronted with Union soldiers. At Gallatin a
camp of instruction, said to have been one of the most beautiful and
extensive in the State, was utterly destroyed. After leaving Gallatin,
the Union cavalry encountered a rebel cavalry force under Colonel
Garland, and a skirmish occurred, in which the enemy suffered
severely in killed, wounded, and prisoners. Two clever feints,
menacing Port Gibson and Natchez, deceived the enemy again, and
the main body marched straight forward to Brookhaven, which was
already occupied by Colonel Grierson’s advance.
On the 30th, the whole force left Brookhaven, and proceeded to
Bogue Cluto Station, destroying every bridge on the way. At the
station, fifteen freight cars, which were standing on the track,
partially loaded, were fired, and utterly destroyed. From that point to
Summit a rapid march was performed, and there twenty-five freight
cars were reduced to ashes. Information was sent to Colonel Grierson
that a regiment of rebel cavalry was rapidly moving toward Wassitta,
and they were discovered at Wall’s bridge on the Tickfaw, by a
detachment of Union cavalry, who dashed in upon them, and killing
and wounding a large number, put the rest to flight. Colonel
Grierson’s loss was one killed and five wounded.
The march again continued, at first east of the Tickfaw, and then
changing again continued directly south, marching through woods,
lanes and by-roads, and struck a road which led directly from Clinton
to Osyka. There the cavalry came most unexpectedly upon the Ninth
Tennessee cavalry regiment, which was posted in a strong defile
guarding the bridges across the Tickfaw river. A sharp skirmish
ensued, in which the enemy’s pickets were captured, and the
regiment driven back with great loss. The cavalry then crossed the
river at Edward’s bridge, where they were met by Garland’s rebel
cavalry, which they put to flight with a single battalion of the Sixth
Illinois, and two guns of the battery, without even halting the
column. It was clearly perceived now that the rebels were sending
out forces in all directions to intercept the march of Colonel
Grierson’s troops.
At midnight the Amite river was crossed, over which there was but
one bridge; and the National troops were just in time to escape a
heavy column of infantry and artillery which had been sent to
intercept them. They moved on to Sandy creek, where Hughes’
cavalry, under Lieutenant-Colonel Wilburn, were encamped, and
reached that point at dawn of day. The rebel camp, completely
surprised, was in no condition to make a successful resistance, and a
large number of the enemy were taken prisoners; the camp,
consisting of one hundred and fifty tents, was destroyed, together
with a great quantity of ammunition, guns, public and private stores,
books, papers, and public documents. A large number of horses were
captured also; and the cavalry then took the road to Baton Rouge,
and on the way surprised Stuart’s cavalry at Commite river, and took
prisoner forty men with their arms and horses.
On the morning of the 1st of May, the commander at Baton Rouge
was startled by the intelligence of the close proximity of Grierson’s
cavalry, and sent out two companies under Captain Godfrey, to meet
and welcome them. The troops entered the city at three o’clock in the
afternoon, amid cheers and shouts of welcome that rent the air, and
echoed along the hills toward Port Hudson. Thus in less than sixteen
days this heroic cavalry force had marched over six hundred miles,
over marshes and rivers, endangering their lives for whole days at
every mile they traversed. The last twenty-eight hours’ march was
performed without either rest or food to men or horses. The loss to
Colonel Grierson’s command during the whole journey was three
killed, seven wounded, five sick and left upon the route, and nine
men missing. Of the enemy, over one hundred were killed and
wounded; five hundred taken prisoners (many of them officers);
from fifty to sixty miles of railroad and telegraph wire destroyed, and
three thousand stand of arms, together with army stores and
government property, captured and destroyed—making in all a loss
to the rebels of over three million dollars.
BATTLE OF RAYMOND, MISS.
May 12, 1863.

On Thursday, May 7th, General McPherson moved his corps to


Rocky Spring, and his camp was occupied next day by General
Sherman. On Saturday McPherson again moved to the eastward, to
the village of Utica, crossing the road occupied by McClernand, and
leaving the latter on his left. On Sunday morning McClernand
marched to Five Mile creek, and encamped on the south bank at
noon, on account of broken bridges, which were repaired the same
day. Monday morning Sherman’s corps came up, passed
McClernand’s, and encamped that night at the village of Auburn,
about ten miles south of Edwards’ Station, on the railroad from
Vicksburg to Jackson. As soon as it passed, McClernand’s corps
followed a few miles, and then took a road going obliquely to the left,
leading to Hall’s Ferry, on the Big Black river. Thus, on Monday
evening General McClernand was at Hall’s Ferry; General Sherman
was at Auburn, six or eight miles to the northeast, and General
McPherson was about eight miles still further to the northeast, a few
miles north of Utica. The whole formed an immense line of battle,
Sherman’s corps being in the centre, with those of McPherson and
McClernand forming the right and left wings. From Grand Gulf the
army marched westward, but, by these last movements, swung on
the left as a pivot, and fronted nearly northward.
Up to this the enemy had not appeared on the line of march. On
Tuesday morning General McClernand’s advance drove in the
enemy’s pickets near Hall’s Ferry, and brisk skirmishing ensued for
an hour or two, with little loss to either side. By noon the rebels had
disappeared from his front, and seven wounded and none killed was
the total Union loss. General Sherman put Steele’s division in motion
early in the morning, and came upon the enemy at the crossing of
Fourteen Mile creek four miles from Auburn. The cavalry advance
was fired into from the thick woods that skirt the stream, but was
unable, owing to the nature of the ground, to make a charge or clear
the rebels from their position. Landgraber’s battery was thrown to
the front, supported by the Seventeenth Missouri and Thirty-first
Iowa infantry regiments, and threw a few shell into the bushy
undergrowth skirting the stream which gave them cover. Skirmishers
were thrown out, and advanced to the creek, driving the enemy
slowly. A brigade was thrown to the right and left flanks, when the
rebel forces, mainly cavalry, withdrew toward Raymond. The bridge
was burned during the skirmish, but a crossing was constructed in
two hours, and trains were passing before noon.
But the principal opposition to the line of march was in the front of
General McPherson. General Logan’s division came upon a body of
rebel troops, estimated at about ten thousand, posted on Fondren’s
creek, two miles southwest of this, at ten o’clock on Tuesday
morning. Brisk skirmishing began at once, and a general engagement
was soon brought on. The enemy (as in front of General Sherman),
was almost wholly concealed at first by the woods bordering the
stream, behind which their forces were posted. Their artillery was on
an eminence that commanded our approach. The Federal troops had
to cross an open field, exposed to a terrible fire. The First and Second
brigades, commanded by General J. E. Smith, and General Fennis
(both Illinois regiments), were in the thickest of the fight, and
suffered most. After three hours’ hard fighting, the enemy withdrew
sullenly in two columns, the principal one taking the road to
Jackson. The Federal loss, in killed, wounded, and missing, was
about three hundred. The rebel loss was much greater.
CAPTURE OF JACKSON, MISS.
May 14, 1863.

On the 13th, General McPherson moved to Clinton, and destroyed


the railroads and telegraph. General Sherman moved to a parallel
position on the Mississippi Springs and Jackson road, and General
McClernand to a point near Raymond.
On the 14th General McPherson and General Sherman each
advanced from his respective position toward Jackson. The rain had
fallen in torrents during the night before, and it continued to fall
until about noon, thus making the roads at first slippery and then
miry. Nevertheless, the troops marched in excellent order and spirits
about fourteen miles, when they came upon the enemy. The main
body of their force in Jackson had marched out on the Clinton road,
and encountered General McPherson about two and a half miles
from the city. A small force of artillery and infantry also took a strong
position in front of General Sherman, about the same distance out
from Jackson.
On the march of General McPherson from Clinton toward Jackson,
General Crocker’s division held the advance. All was quiet until he
reached a hill overlooking a broad open field, through the centre of
which and over the crest of the hill beyond, the road to Jackson
passed. On the left of this latter hill the enemy had posted his
artillery, and along the crest his line of battle. As the Federal force
came within range, the artillery of the enemy opened fire. The
battery of the First Missouri was moved to the left of a cotton gin in
the open field, and returned the fire for nearly an hour, when the
guns of the enemy were withdrawn. Meantime, General Crocker had
thrown out two brigades to the right and left of his battery,
supported by another brigade at a proper distance, and had also
pushed forward a strong line of skirmishers, and posted them in a
ravine in front, which protected them from the fire of the enemy.
After a little delay they were again advanced out of cover, and a
desultory fire ensued between the opposite line of skirmishers, in
which the enemy, owing to the nature of the ground, had the
advantage. At length General Crocker, seeing the necessity of driving
the rebels from the crest of the hill, ordered a charge along the line.
With colors flying, and with a step as measured and unbroken as if
on dress parade, the movement was executed. Slowly they advanced,
crossed the narrow ravine, and, with fixed bayonets, reached the
crest of the hill in easy range of the rebel line. Here they received a
tremendous volley, which caused painful gaps in their ranks. They
held their fire until they were within a distance of thirty paces, when
they delivered the returning volley with fearful effect, and, without
waiting to reload their muskets, with a terrific yell, they rushed upon
the staggered foe. Over the fences, through the brushwood into the
enclosure, they worked their way, slaughtering on the right and left
without mercy. The enemy, astonished at their impetuosity, wavered
and fell back, rallied again, and finally broke in wild confusion. They
finally retreated north, but without further damage.
When General Sherman encountered the enemy, he discovered
their weakness by sending a reconnoitering party to his right, which
had the effect of causing them to retreat from that part of their line.
A few artillerists, however, remained in their places, firing upon
General Sherman’s troops until the last moment.
At this time General McClernand occupied Clinton with one
division, Mississippi Springs with another, Raymond with a third,
and his fourth division and General Blair’s division of General
Sherman’s corps were with a wagon train, still in the rear, near
Auburn. At the same time General McArthur, with one brigade of his
division of General McPherson’s corps, was moving toward Raymond
on the Utica road. It was not the intention of General Grant to move
these forces any nearer Jackson, but to have them in a position
where they could be in supporting distance if the resistance at
Jackson should prove more obstinate than there seemed any reason
to expect.
On the retreat of the enemy, General McPherson followed directly
into the city of Jackson. A fine battery of six pieces was found, and
around the Deaf and Dumb Institute, which was used as a hospital,
tents enough were seized to encamp an entire division. The
commissary and quartermaster’s stores were in flames. The
Governor and State Treasurer had withdrawn, taking the State funds
and papers. All citizens officially connected with the State or
Confederate Governments had also left. Many soldiers remained,
besides a large number in the hospital.
Early on the morning of the day following the occupation of the
city of Jackson it was decided to evacuate the position. There were
several reasons which induced General Grant to arrive at this
decision, prominent among which was the difficulty of keeping intact
his long line of communication, and the fear that General Johnston—
who was known to be hovering in the region north of Jackson with a
force estimated at twenty thousand men—would attack his rear. The
force which he encountered just before reaching Jackson, under
General Gregg, had divided, one portion going to Canton from the
north, and the other from the south. Johnston and Gregg combined
might prove altogether too formidable. It was therefore decided to
return to Clinton and move upon Vicksburg.
The main column of the enemy was at Edwards’ Station, proposing
to give battle there. Soon after daylight the column was in motion,
General McPherson in advance. They reached Clinton at noon, and
after an hour’s delay marched to their camping ground, a short
distance from the village of Bolton.
The programme of the advance was arranged by General Grant
and General McClernand as follows:—Extreme left, General Smith,
supported by General Blair; on the right of General Smith, General
Osterhaus, supported by General Carr; General Hovey in the centre,
with McPherson on the extreme right, and Crocker as reserve. In this
order the advance was made; General McClernand’s corps, with the
exception of General Hovey’s division, reaching the position by way
of the several roads leading from Raymond to Edward’s Station.
On the evening of the 15th, General McClernand heard that the
enemy was advancing from Edwards’ Station to Raymond, and
quickly placed his troops in order of battle to repel the anticipated
attack. Extensive reconnoissances revealed the fact, however, that he
was merely feeling his position and force, and that no attack need be
expected that day.
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!

ebookname.com

You might also like