Python Data Science

ISBN-13: 9789811977015 / Angielski / Twarda / 2023 / 1376 str.

Chaolemen Borjigin

Opis

Rather than presenting Python as Java or C, this textbook focuses on the essential Python programming skills for data scientists and advanced methods for big data analysts.Unlike conventional textbooks, it is based on Markdown and uses full-color printing and a code-centric approach to highlight the 3C principles in data science: creative design of data solutions, curiosity about the data lifecycle, and critical thinking regarding data insights. Q&A-based knowledge maps, tips and suggestions, notes, as well as warnings and cautions are employed to explain the key points, difficulties, and common mistakes in Python programming for data science. In addition, it includes suggestions for further reading. This textbook provides an open-source community via GitHub, and the course materials are licensed for free use under the following license: Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0).

Spis treści

1. Python and Data Science

Q&A

1.1 From data analysis to data science

1.2 Python language and its characteristics

1.3 Precautions for data analysis based on Python

1.4 Python development environment and how to build it

  Exercises

 

2. Basic Python Programming for Data Science

Q&A

2.1 Variables and their definition methods

2.2 Operators, expressions, statements

2.3 Data type and data structure

2.4 Packages and modules

2.5 Built-in functions, module functions and custom functions

Exercises

 

3. Advanced Python Programming for Data Science

Q&A

3.1 Iterators and iterable objects

3.2 Decorators and generators

3.3 Help and Doc Strings

3.4 Exception handling, assertion and debugging

3.5 Search path, current working directory

3.6 Object-oriented programming

Exercises

 

4. Data preprocessing and wrangling

Q&A

4.1 Random numbers and Random/Sklearn

4.2 Vectorized computing and NumPy

4.3 Data frame calculation and Pandas

4.4 Data visualization and MatPlotlib/Seaborn and others

Exercises

 

5. Data analysis algorithms and models

Q&A

5.1 Statistical modelling with statsmodels

5.2 Machine learning with scikit-learn

Exercises

 

O autorze

Chaolemen Borjigin is an associate professor at Renmin University of China, and one of the top 50 data science influencers in China. He is a member of the Information System Special Committee of the Chinese Computer Federation, deputy director of the Expert Committee of the National University Artificial Intelligence and Big Data Innovation Alliance of China, executive editorial board member of the academic journal Computer Science, and deputy editor-in-chief of the international journal Data Science and Informatics.

He is the author of Data Science (Tsinghua University Press, 2016), the first monograph in China that systematically introduced data science principles, theories, methods, technologies, and tools. His textbook Data Science Theory and Practice (Second Edition) was recognized as a high-quality textbook by the Beijing Municipal Education Commission in 2019. His course Introduction to Data Science is one of the China National First-Class Undergraduate Courses.

Blurb

Rather than presenting Python as Java or C, this book focuses on the essential Python programming skills for data scientists and advanced methods for big data analysts.

Unlike conventional textbooks, it is based on Markdown and uses full-color printing and a code-centric approach to highlight the 3C principles in data science: creative design of data solutions, curiosity about the data lifecycle, and critical thinking regarding data insights. Q&A-based knowledge maps, tips and suggestions, notes, as well as warnings and cautions are employed to explain the key points, difficulties, and common mistakes in Python programming for data science. In addition, it includes suggestions for further reading. 

This textbook provides an open-source community via GitHub, and the course materials are licensed for free use under the following license: Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0).