Whether you are new to scientific/analytic programming, or a seasoned expert, this book will provide you with the skills you need to successfully create, optimize, and distribute your Python/NumPy analytical modules.
Starting from the beginning, this book will cover the key features of NumPy arrays and the details of tuning the data format to make it most fit to your analytical needs. You will then get a walkthrough of the core and submodules that are common to various multidimensional, data-typed analysis. Next, you will move on to key technical implementations, such as linear algebra and Fourier analysis. Finally, you will learn about extending your NumPy capabilities for both functionality and performance by using Cython and the NumPy C API. The last chapter of this book also provides advanced materials to help you learn further by yourself.
This guide is an invaluable tutorial if you are planning to use NumPy in analytical projects.
Chapter 1, An Introduction to NumPy, is a Getting Started chapter of this book, which provides the instructions to help you set up the environment. It starts with introducing the Scientific Python Module family (SciPy Stack) and explains the key role NumPy plays in scientific computing with Python.
Chapter 2, The NumPy ndarray Object, covers the essential usage of NumPy ndarray object, including the initialization, the fundamental attributes, data types, and memory layout. It also covers the theory underneath the operation, which gives you a clear picture of ndarray.
Chapter 3, Using Numpy Arrays, is an advanced chapter on NumPy ndarray usage, which continues Chapter 2, The NumPy ndarray Object. It covers the universal functions in NumPy and shows you the tricks to speed up your code. It also shows you the shape manipulation and broadcasting rules.
Chapter 4, Numpy Core and Libs Submodules, includes two sections. The first section has detailed explanation about the relationship between the way NumPy ndarray allocates memory and the interaction of CPU cache. The second part of this chapter covers the special NumPy Array containing multiple data types (the structure/record array). Also, this chapter explores the experimental datetime64 module in NumPy.
Chapter 5, Linear Algebra in NumPy, starts by utilizing matrix and mathematical computation using linear algebra modules. It shows you multiple ways to solve a mathematical problem: using Matrix, vector decomposition, and polynomials. It also provides concrete practice for curve fitting and regression.
Chapter 6, Fourier Analysis in NumPy, covers the signal processing with NumPy FFT module and the Fourier application on amplifying signals/enlarging images without distortion. It also provides the basic usage of the matplotlib package in Python.
Chapter 7, Building and Distributing NumPy Code, covers the basic details around packaging and publishing the code in Python. It provides a basic introduction to NumPy-specific setup files and how to build extension modules.
Chapter 8, Speeding Up NumPy with Cython, introduces the users to the Cython programming language and introduces readers to techniques that can be used to speed up existing Python code.
Chapter 9, Introduction to the NumPy C-API, provides a basic introduction to the NumPy C API and, in general, how to write wrappers around the existing C/C++ library. The chapter aims to provide a gentle introduction along with equipping the readers with a basic knowledge of how to create new wrappers and understand the existing programs.
Chapter 10, Further Reading, is the last chapter of this book. It gives a summary of what we've learned in the book and explores 4 SciPy stack Python modules relying on NumPy arrays, which give you ideas about further scientific Python programming.