Learning pandas(Second Edition)
上QQ阅读APP看书,第一时间看更新

Numeric and scientific computing - NumPy and SciPy

NumPy (http://www.numpy.org/) is the cornerstone toolbox for scientific computing with Python, and is included in most distributions of modern Python. It is actually a foundational toolbox from which pandas was built, and when using pandas you will almost certainly use it frequently. NumPy provides, among other things, support for multidimensional arrays with basic operations on them and useful linear algebra functions.

The use of the array features of NumPy goes hand in hand with pandas, specifically the pandas Series object. Most of our examples will reference NumPy, but the pandas Series functionality is such a tight superset of the NumPy array that we will, except for a few brief situations, not delve into details of NumPy.

SciPy (https://www.scipy.org/) provides a collection of numerical algorithms and domain-specific toolboxes, including signal processing, optimization, statistics, and much more.