Book Image

Learn Python by Building Data Science Applications

By : Philipp Kats, David Katz
Book Image

Learn Python by Building Data Science Applications

By: Philipp Kats, David Katz

Overview of this book

Python is the most widely used programming language for building data science applications. Complete with step-by-step instructions, this book contains easy-to-follow tutorials to help you learn Python and develop real-world data science projects. The “secret sauce” of the book is its curated list of topics and solutions, put together using a range of real-world projects, covering initial data collection, data analysis, and production. This Python book starts by taking you through the basics of programming, right from variables and data types to classes and functions. You’ll learn how to write idiomatic code and test and debug it, and discover how you can create packages or use the range of built-in ones. You’ll also be introduced to the extensive ecosystem of Python data science packages, including NumPy, Pandas, scikit-learn, Altair, and Datashader. Furthermore, you’ll be able to perform data analysis, train models, and interpret and communicate the results. Finally, you’ll get to grips with structuring and scheduling scripts using Luigi and sharing your machine learning models with the world as a microservice. By the end of the book, you’ll have learned not only how to implement Python in data science projects, but also how to maintain and design them to meet high programming standards.
Table of Contents (26 chapters)
Free Chapter
1
Section 1: Getting Started with Python
11
Section 2: Hands-On with Data
17
Section 3: Moving to Production

Getting started with pandas

Pandas is the tool for data manipulation in Python—it combines speed and convenience, allowing the rapid processing and manipulation of data. Let's first overview a number of basic operations: pandas is simple and intuitive to use, but it is still a learning curve.

pandas does have two main data structures:

  1. Series is a one-dimensional array of one data type that also has an index. The index could be numeric, categorical, a string, or datetime.
  2. DataFrame is a two-dimensional table consisting of a set of columns—each of one single data type. Dataframe has two indexes—index and columns. Columns of Dataframe can be thought of as Series. Rows can be retrieved as Series but, in this case, data in the cells will likely be converted to one shared data type object (more on that later).

Most of the time, we get our data from external...