Python for Data Analysis : Data Wrangling with Pandas, NumPy, and IPython / Wes McKinney,

All Indian Reprints of O'Reilly are printed in Grayscale

Get complete instructions for manipulating, processing, cleaning, and crunching datasets in Python. Updated for Python 3.6, the second edition of this hands-on guide is packed with practical case studies that show you how to solve a broad set of data analysis problems effectively. You'll learn the latest versions of pandas, NumPy, IPython, and Jupyter in the process.Written by Wes McKinney, the creator of the Python pandas project, this book is a practical, modern introduction to data science tools in Python. It's ideal for analysts new to Python and for Python programmers new to data science and scientific computing. Data files and related material are available on GitHub.

Use the IPython shell and Jupyter notebook for exploratory computingLearn basic and advanced features in NumPy (Numerical Python)Get started with data analysis tools in the pandas libraryUse flexible tools to load, clean, transform, merge, and reshape dataCreate informative visualizations with matplotlibApply the pandas groupby facility to slice, dice, and summarize datasetsAnalyze and manipulate regular and irregular time series dataLearn how to solve real-world data analysis problems with thorough, detailed examples

About the Author

Wes McKinney is a New York−based software developer and entrepreneur. After finishing his undergraduate degree in mathematics at MIT in 2007, he went on to do quantitative finance work at AQR Capital Management in Greenwich, CT. Frustrated by cumbersome data analysis tools, he learned Python and started building what would later become the pandas project. He's now an active member of the Python data community and is an advocate for the use of Python in data analysis, finance, and statistical computing applications.

Wes was later the co-founder and CEO of DataPad, whose technology assets and team were acquired by Cloudera in 2014. He has since become involved in big data technology, joining the Project Management Committees for the Apache Arrow and Apache Parquet projects in the Apache Software Foundation. In 2016, he joined Two Sigma Investments in New York City, where he continues working to make data analysis faster and easier through open source software.

TABLE OF CONTENTS
Preface. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi
1. Preliminaries. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1 What Is This Book About? 1
What Kinds of Data? 1
1.2 Why Python for Data Analysis? 2
Python as Glue 2
Solving the “Two-Language” Problem 3
Why Not Python? 3
1.3 Essential Python Libraries 4
NumPy 4
pandas 4
matplotlib 5
IPython and Jupyter 6
SciPy 6
scikit-learn 7
statsmodels 8
1.4 Installation and Setup 8
Windows 9
Apple (OS X, macOS) 9
GNU/Linux 9
Installing or Updating Python Packages 10
Python 2 and Python 3 11
Integrated Development Environments (IDEs) and Text Editors 11
1.5 Community and Conferences 12
1.6 Navigating This Book 12
Code Examples 13
Data for Examples 13
iii
Import Conventions 14
Jargon 14
2. Python Language Basics, IPython, and Jupyter Notebooks. . . . . . . . . . . . . . . . . . . . . . . . 15
2.1 The Python Interpreter 16
2.2 IPython Basics 17
Running the IPython Shell 17
Running the Jupyter Notebook 18
Tab Completion 21
Introspection 23
The %run Command 25
Executing Code from the Clipboard 26
Terminal Keyboard Shortcuts 27
About Magic Commands 28
Matplotlib Integration 29
2.3 Python Language Basics 30
Language Semantics 30
Scalar Types 38
Control Flow 46
3. Built-in Data Structures, Functions, and Files. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
3.1 Data Structures and Sequences 51
Tuple 51
List 54
Built-in Sequence Functions 59
dict 61
set 65
List, Set, and Dict Comprehensions 67
3.2 Functions 69
Namespaces, Scope, and Local Functions 70
Returning Multiple Values 71
Functions Are Objects 72
Anonymous (Lambda) Functions 73
Currying: Partial Argument Application 74
Generators 75
Errors and Exception Handling 77
3.3 Files and the Operating System 80
Bytes and Unicode with Files 83
3.4 Conclusion 84
4. NumPy Basics: Arrays and Vectorized Computation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
4.1 The NumPy ndarray: A Multidimensional Array Object 87
iv | Table of Contents
Creating ndarrays 88
Data Types for ndarrays 90
Arithmetic with NumPy Arrays 93
Basic Indexing and Slicing 94
Boolean Indexing 99
Fancy Indexing 102
Transposing Arrays and Swapping Axes 103
4.2 Universal Functions: Fast Element-Wise Array Functions 105
4.3 Array-Oriented Programming with Arrays 108
Expressing Conditional Logic as Array Operations 109
Mathematical and Statistical Methods 111
Methods for Boolean Arrays 113
Sorting 113
Unique and Other Set Logic 114
4.4 File Input and Output with Arrays 115
4.5 Linear Algebra 116
4.6 Pseudorandom Number Generation 118
4.7 Example: Random Walks 119
Simulating Many Random Walks at Once 121
4.8 Conclusion 122
5. Getting Started with pandas. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
5.1 Introduction to pandas Data Structures 124
Series 124
DataFrame 128
Index Objects 134
5.2 Essential Functionality 136
Reindexing 136
Dropping Entries from an Axis 138
Indexing, Selection, and Filtering 140
Integer Indexes 145
Arithmetic and Data Alignment 146
Function Application and Mapping 151
Sorting and Ranking 153
Axis Indexes with Duplicate Labels 157
5.3 Summarizing and Computing Descriptive Statistics 158
Correlation and Covariance 160
Unique Values, Value Counts, and Membership 162
5.4 Conclusion 165
6. Data Loading, Storage, and File Formats. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167
6.1 Reading and Writing Data in Text Format 167
Table of Contents | v
Reading Text Files in Pieces 173
Writing Data to Text Format 175
Working with Delimited Formats 176
JSON Data 178
XML and HTML: Web Scraping 180
6.2 Binary Data Formats 183
Using HDF5 Format 184
Reading Microsoft Excel Files 186
6.3 Interacting with Web APIs 187
6.4 Interacting with Databases 188
6.5 Conclusion 190
7. Data Cleaning and Preparation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191
7.1 Handling Missing Data 191
Filtering Out Missing Data 193
Filling In Missing Data 195
7.2 Data Transformation 197
Removing Duplicates 197
Transforming Data Using a Function or Mapping 198
Replacing Values 200
Renaming Axis Indexes 201
Discretization and Binning 203
Detecting and Filtering Outliers 205
Permutation and Random Sampling 206
Computing Indicator/Dummy Variables 208
7.3 String Manipulation 211
String Object Methods 211
Regular Expressions 213
Vectorized String Functions in pandas 216
7.4 Conclusion 219
8. Data Wrangling: Join, Combine, and Reshape. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 221
8.1 Hierarchical Indexing 221
Reordering and Sorting Levels 224
Summary Statistics by Level 225
Indexing with a DataFrame’s columns 225
8.2 Combining and Merging Datasets 227
Database-Style DataFrame Joins 227
Merging on Index 232
Concatenating Along an Axis 236
Combining Data with Overlap 241
8.3 Reshaping and Pivoting 242
vi | Table of Contents
Reshaping with Hierarchical Indexing 243
Pivoting “Long” to “Wide” Format 246
Pivoting “Wide” to “Long” Format 249
8.4 Conclusion 251
9. Plotting and Visualization. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 253
9.1 A Brief matplotlib API Primer 253
Figures and Subplots 255
Colors, Markers, and Line Styles 259
Ticks, Labels, and Legends 261
Annotations and Drawing on a Subplot 265
Saving Plots to File 267
matplotlib Configuration 268
9.2 Plotting with pandas and seaborn 268
Line Plots 269
Bar Plots 272
Histograms and Density Plots 277
Scatter or Point Plots 280
Facet Grids and Categorical Data 283
9.3 Other Python Visualization Tools 285
9.4 Conclusion 286
10. Data Aggregation and Group Operations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 287
10.1 GroupBy Mechanics 288
Iterating Over Groups 291
Selecting a Column or Subset of Columns 293
Grouping with Dicts and Series 294
Grouping with Functions 295
Grouping by Index Levels 295
10.2 Data Aggregation 296
Column-Wise and Multiple Function Application 298
Returning Aggregated Data Without Row Indexes 301
10.3 Apply: General split-apply-combine 302
Suppressing the Group Keys 304
Quantile and Bucket Analysis 305
Example: Filling Missing Values with Group-Specific Values 306
Example: Random Sampling and Permutation 308
Example: Group Weighted Average and Correlation 310
Example: Group-Wise Linear Regression 312
10.4 Pivot Tables and Cross-Tabulation 313
Cross-Tabulations: Crosstab 315
10.5 Conclusion 316
Table of Contents | vii
11. Time Series. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 317
11.1 Date and Time Data Types and Tools 318
Converting Between String and Datetime 319
11.2 Time Series Basics 322
Indexing, Selection, Subsetting 323
Time Series with Duplicate Indices 326
11.3 Date Ranges, Frequencies, and Shifting 327
Generating Date Ranges 328
Frequencies and Date Offsets 330
Shifting (Leading and Lagging) Data 332
11.4 Time Zone Handling 335
Time Zone Localization and Conversion 335
Operations with Time Zone−Aware Timestamp Objects 338
Operations Between Different Time Zones 339
11.5 Periods and Period Arithmetic 339
Period Frequency Conversion 340
Quarterly Period Frequencies 342
Converting Timestamps to Periods (and Back) 344
Creating a PeriodIndex from Arrays 345
11.6 Resampling and Frequency Conversion 348
Downsampling 349
Upsampling and Interpolation 352
Resampling with Periods 353
11.7 Moving Window Functions 354
Exponentially Weighted Functions 358
Binary Moving Window Functions 359
User-Defined Moving Window Functions 361
11.8 Conclusion 362
12. Advanced pandas. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 363
12.1 Categorical Data 363
Background and Motivation 363
Categorical Type in pandas 365
Computations with Categoricals 367
Categorical Methods 370
12.2 Advanced GroupBy Use 373
Group Transforms and “Unwrapped” GroupBys 373
Grouped Time Resampling 377
12.3 Techniques for Method Chaining 378
The pipe Method 380
12.4 Conclusion 381
viii | Table of Contents
13. Introduction to Modeling Libraries in Python. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 383
13.1 Interfacing Between pandas and Model Code 383
13.2 Creating Model Descriptions with Patsy 386
Data Transformations in Patsy Formulas 389
Categorical Data and Patsy 390
13.3 Introduction to statsmodels 393
Estimating Linear Models 393
Estimating Time Series Processes 396
13.4 Introduction to scikit-learn 397
13.5 Continuing Your Education 401
14. Data Analysis Examples. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 403
14.1 1.USA.gov Data from Bitly 403
Counting Time Zones in Pure Python 404
Counting Time Zones with pandas 406
14.2 MovieLens 1M Dataset 413
Measuring Rating Disagreement 418
14.3 US Baby Names 1880–2010 419
Analyzing Naming Trends 425
14.4 USDA Food Database 434
14.5 2012 Federal Election Commission Database 440
Donation Statistics by Occupation and Employer 442
Bucketing Donation Amounts 445
Donation Statistics by State 447
14.6 Conclusion 448
A. Advanced NumPy. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 449
A.1 ndarray Object Internals 449
NumPy dtype Hierarchy 450
A.2 Advanced Array Manipulation 451
Reshaping Arrays 452
C Versus Fortran Order 454
Concatenating and Splitting Arrays 454
Repeating Elements: tile and repeat 457
Fancy Indexing Equivalents: take and put 459
A.3 Broadcasting 460
Broadcasting Over Other Axes 462
Setting Array Values by Broadcasting 465
A.4 Advanced ufunc Usage 466
ufunc Instance Methods 466
Writing New ufuncs in Python 468
A.5 Structured and Record Arrays 469
Table of Contents | ix
Nested dtypes and Multidimensional Fields 469
Why Use Structured Arrays? 470
A.6 More About Sorting 471
Indirect Sorts: argsort and lexsort 472
Alternative Sort Algorithms 474
Partially Sorting Arrays 474
numpy.searchsorted: Finding Elements in a Sorted Array 475
A.7 Writing Fast NumPy Functions with Numba 476
Creating Custom numpy.ufunc Objects with Numba 478
A.8 Advanced Array Input and Output 478
Memory-Mapped Files 478
HDF5 and Other Array Storage Options 480
A.9 Performance Tips 480
The Importance of Contiguous Memory 480
B. More on the IPython System. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 483
B.1 Using the Command History 483
Searching and Reusing the Command History 483
Input and Output Variables 484
B.2 Interacting with the Operating System 485
Shell Commands and Aliases 486
Directory Bookmark System 487
B.3 Software Development Tools 487
Interactive Debugger 488
Timing Code: %time and %timeit 492
Basic Profiling: %prun and %run -p 494
Profiling a Function Line by Line 496
B.4 Tips for Productive Code Development Using IPython 498
Reloading Module Dependencies 498
Code Design Tips 499
B.5 Advanced IPython Features 500
Making Your Own Classes IPython-Friendly 500
Profiles and Configuration 501
B.6 Conclusion 503
Index. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 505
x | Table of Contents