Python Data Analysis: A Beginner's Guide to NumPy, 2nd Edition

This is an introductory NumPy tutorial for beginners. NumPy is the foundational library of Python's scientific computing ecosystem, providing multidimensional array objects, a large set of mathematical functions and efficient vectorized operations; it can replace many common features of Matlab and Mathematica and is free and open source. The book starts with installation: configuring Python, NumPy, Matplotlib, SciPy and IPython on different operating systems, building a first NumPy program through a small example such as vector addition, then systematically covering array objects and data types, indexing and slicing, reshaping and combining/splitting, reading and writing files and CSVs, and common statistical and financial functions (volume-weighted average price, moving averages, Bollinger bands and so on, all illustrated with stock data). It then moves to advanced material: convenience functions and polynomial fitting, matrices and universal functions (ufuncs), the linear algebra module (inverses, solving linear systems, eigenvalues, singular value decomposition, determinants), fast Fourier transforms and random number distributions, specialized functions (sorting, searching, financial functions, window functions, special mathematical functions), writing unit tests for NumPy code with the assert family of functions, plotting with Matplotlib, and integrated applications of SciPy, Scikits and Pygame. Sections mostly follow a "Time for Action" and "What Just Happened" structure with short examples and explanations, suited to typing along as you read. The Chinese edition has 11 chapters and 226 pages.