Pandas for Everyone: Python Data Analysis / Edition 1

by Daniel Chen

Overview

The Hands-On, Example-Rich Introduction to Pandas Data Analysis in Python

Today, analysts must manage data characterized by extraordinary variety, velocity, and volume. Using the open source Pandas library, you can use Python to rapidly automate and perform virtually any data analysis task, no matter how large or complex. Pandas can help you ensure the veracity of your data, visualize it for effective decision-making, and reliably reproduce analyses across multiple datasets.

Pandas for Everyone brings together practical knowledge and insight for solving real problems with Pandas, even if you’re new to Python data analysis. Daniel Y. Chen introduces key concepts through simple but practical examples, incrementally building on them to solve more difficult, real-world problems.

Chen gives you a jumpstart on using Pandas with a realistic dataset and covers combining datasets, handling missing data, and structuring datasets for easier analysis and visualization. He demonstrates powerful data cleaning techniques, from basic string manipulation to applying functions simultaneously across dataframes.

Once your data is ready, Chen guides you through fitting models for prediction, clustering, inference, and exploration. He provides tips on performance and scalability, and introduces you to the wider Python data analysis ecosystem.

• Work with DataFrames and Series, and import or export data
• Create plots with matplotlib, seaborn, and pandas
• Combine datasets and handle missing data
• Reshape, tidy, and clean datasets so they’re easier to work with
• Convert data types and manipulate text strings
• Apply functions to scale data manipulations
• Aggregate, transform, and filter large datasets with groupby
• Leverage Pandas’ advanced date and time capabilities
• Fit linear models using statsmodels and scikit-learn libraries
• Use generalized linear modeling to fit models with different response variables
• Compare multiple models to select the “best”
• Regularize to overcome overfitting and improve performance
• Use clustering in unsupervised machine learning

Product Details

ISBN-13:	9780134546933
Publisher:	Pearson Education
Publication date:	12/26/2017
Series:	Addison-Wesley Data & Analytics Series
Pages:	416
Product dimensions:	6.90(w) x 9.00(h) x 1.10(d)

About the Author

Daniel Chen is a graduate student in the interdisciplinary PhD program in Genetics, Bioinformatics & Computational Biology (GBCB) at Virginia Tech. He is involved with Software Carpentry as an instructor and lesson maintainer. He completed his master’s degree in public health at Columbia University Mailman School of Public Health in Epidemiology, and currently works at the Social and Decision Analytics Laboratory under the Biocomplexity Institute of Virginia Tech where he is working with data to inform policy decision-making. He is the author of Pandas for Everyone and Pandas Data Analysis with Python Fundamentals LiveLessons.

Table of Contents

Foreword xix

Preface xxi

Acknowledgments xxvii

About the Author xxxi

Part I: Introduction 1

Chapter 1: Pandas DataFrame Basics 3

1.1 Introduction 3

1.2 Loading Your First Data Set 4

1.3 Looking at Columns, Rows, and Cells 7

1.4 Grouped and Aggregated Calculations 18

1.5 Basic Plot 23

1.6 Conclusion 24

Chapter 2: Pandas Data Structures 25

2.1 Introduction 25

2.2 Creating Your Own Data 26

2.3 The Series 28

2.4 The DataFrame 36

2.5 Making Changes to Series and DataFrames 38

2.6 Exporting and Importing Data 43

2.7 Conclusion 47

Chapter 3: Introduction to Plotting 49

3.1 Introduction 49

3.2 Matplotlib 51

3.3 Statistical Graphics Using matplotlib 56

3.4 Seaborn 61

3.5 Pandas Objects 83

3.6 Seaborn Themes and Styles 86

3.7 Conclusion 90

Part II: Data Manipulation 91

Chapter 4: Data Assembly 93

4.1 Introduction 93

4.2 Tidy Data 93

4.3 Concatenation 94

4.4 Merging Multiple Data Sets 102

4.5 Conclusion 107

Chapter 5: Missing Data 109

5.1 Introduction 109

5.2 What Is a NaN Value? 109

5.3 Where Do Missing Values Come From? 111

5.4 Working with Missing Data 116

5.5 Conclusion 121

Chapter 6: Tidy Data 123

6.1 Introduction 123

6.2 Columns Contain Values, Not Variables 124

6.3 Columns Contain Multiple Variables 128

6.4 Variables in Both Rows and Columns 133

6.5 Multiple Observational Units in a Table (Normalization) 134

6.6 Observational Units Across Multiple Tables 137

6.7 Conclusion 141

Part III: Data Munging 143

Chapter 7: Data Types 145

7.1 Introduction 145

7.2 Data Types 145

7.3 Converting Types 146

7.4 Categorical Data 152

7.5 Conclusion 153

Chapter 8: Strings and Text Data 155

8.1 Introduction 155

8.2 Strings 155

8.3 String Methods 158

8.4 More String Methods 160

8.5 String Formatting 161

8.6 Regular Expressions (RegEx) 164

8.7 The regex Library 170

8.8 Conclusion 170

Chapter 9: Apply 171

9.1 Introduction 171

9.2 Functions 171

9.3 Apply (Basics) 172

9.4 Apply (More Advanced) 177

9.5 Vectorized Functions 182

9.6 Lambda Functions 185

9.7 Conclusion 187

Chapter 10: Groupby Operations: Split–Apply–Combine 189

10.1 Introduction 189

10.2 Aggregate 190

10.3 Transform 197

10.4 Filter 201

10.5 The pandas.core.groupby.DataFrameGroupBy Object 202

10.6 Working with a MultiIndex 207

10.7 Conclusion 211

Chapter 11: The datetime Data Type 213

11.1 Introduction 213

11.2 Python’s datetime Object 213

11.3 Converting to datetime 214

11.4 Loading Data That Include Dates 217

11.5 Extracting Date Components 217

11.6 Date Calculations and Timedeltas 220

11.7 Datetime Methods 221

11.8 Getting Stock Data 224

11.9 Subsetting Data Based on Dates 225

11.10 Date Ranges 227

11.11 Shifting Values 230

11.12 Resampling 237

11.13 Time Zones 238

11.14 Conclusion 240

Part IV: Data Modeling 241

Chapter 12: Linear Models 243

12.1 Introduction 243

12.2 Simple Linear Regression 243

12.3 Multiple Regression 247

12.4 Keeping Index Labels From sklearn 251

12.5 Conclusion 252

Chapter 13: Generalized Linear Models 253

13.1 Introduction 253

13.2 Logistic Regression 253

13.3 Poisson Regression 257

13.4 More Generalized Linear Models 260

13.5 Survival Analysis 260

13.6 Conclusion 264

Chapter 14: Model Diagnostics 265

14.1 Introduction 265

14.2 Residuals 265

14.3 Comparing Multiple Models 270

14.4 k-Fold Cross-Validation 275

14.5 Conclusion 278

Chapter 15: Regularization 279

15.1 Introduction 279

15.2 Why Regularize? 279

15.3 LASSO Regression 281

15.4 Ridge Regression 283

15.5 Elastic Net 285

15.6 Cross-Validation 287

15.7 Conclusion 289

Chapter 16: Clustering 291

16.1 Introduction 291

16.2 k-Means 291

16.3 Hierarchical Clustering 297

16.4 Conclusion 301

Part V: Conclusion 303

Chapter 17: Life Outside of Pandas 305

17.1 The (Scientific) Computing Stack 305

17.2 Performance 306

17.3 Going Bigger and Faster 307

Chapter 18: Toward a Self-Directed Learner 309

18.1 It’s Dangerous to Go Alone! 309

18.2 Local Meetups 309

18.3 Conferences 309

18.4 The Internet 310

18.5 Podcasts 310

18.6 Conclusion 311

Part VI: Appendixes 313

Appendix A: Installation 315

A.1 Installing Anaconda 315

A.2 Uninstall Anaconda 316

Appendix B: Command Line 317

B.1 Installation 317

B.2 Basics 318

Appendix C: Project Templates 319

Appendix D: Using Python 321

D.1 Command Line and Text Editor 321

D.2 Python and IPython 322

D.3 Jupyter 322

D.4 Integrated Development Environments (IDEs) 322

Appendix E: Working Directories 325

Appendix F: Environments 327

Appendix G: Install Packages 329

G.1 Updating Packages 330

Appendix H: Importing Libraries 331

Appendix I: Lists 333

Appendix J: Tuples 335

Appendix K: Dictionaries 337

Appendix L: Slicing Values 339

Appendix M: Loops 341

Appendix N: Comprehensions 343

Appendix O: Functions 345

O.1 Default Parameters 347

O.2 Arbitrary Parameters 347

Appendix P: Ranges and Generators 349

Appendix Q: Multiple Assignment 351

Appendix R: numpy ndarray 353

Appendix S: Classes 355

Appendix T: Odo: The Shapeshifter 357

Index 359