About this project

This repository is the Chinese translation of the second edition of "Python for Data Analysis" by Wes McKinney. The original book was officially published in October 2017 and is a classic introductory textbook in the field of Python data analysis. The Chinese version was translated by the translator in about three months of spare time. Compared with the first edition, the main changes in the second edition include: all code in the book has been upgraded to Python 3.6 (the first edition was based on Python 2.7); the installation methods for Anaconda and other packages have been updated; pandas has been updated to the latest version as of 2017; a new chapter has been added introducing more advanced pandas tools and usage tips; and a brief introduction to StatsModels and scikit-learn has been included. In addition, the Python tutorial in the appendix of the first edition has been split and expanded into Chapters 2 and 3, and some content has been re-typeset. The repository content mainly consists of the Chinese translation of the book chapters. Readers can read the full text online via the Jianshu link provided in the README. The accompanying example code is hosted in the original author's GitHub repository (the 2nd-edition branch of wesm/pydata-book). The translator recommends downloading the code, installing Anaconda, and opening and running it with Jupyter Notebook in the code directory. The translator also maintains a third-edition translation and accompanying study notes repository (python_for_data_analysis_3rd_study_note), which includes chapter-by-chapter guide videos (on Bilibili), a WeChat Reading online reading link, and an errata document compiled from reader feedback. According to the README, the third edition adds about 41 pages compared with the second edition, upgrades pandas to 1.4.0 and Python to 3.10. The main changes include: Chapter 4 on NumPy basics adds pseudo-random number generation; Chapter 7 on data cleaning adds extended data types and categorical data; Chapter 11 on time series adds grouped time resampling; and in programming details, some shorthand usages that could cause memory confusion have been abandoned (e.g., using axis="columns" instead of axis=1), making it more beginner-friendly. It should be noted that the README contains promotional links to a third-party proxy IP service and introductions to the translator's other works (such as the authoritative guide to Polars). These are promotional content and are unrelated to the translation content of this book.