As we said in the first section of this course, a major benefit of using Python for data science in comparison to other programming languages is the availability of a large number of useful packages that are distributed under a free license. There are a few packages in particular that have become so synonymous with Python, they might as well be included alongside the default packages. In fact, Anaconda, a popular distribution of data science packages in one installation, includes all of the packages we will talk about in this section. These packages are:

• numpy, for linear algebra and matrix operations.
• scipy, for computational calculus and statistics.
• matplotlib, for graphical displays
• pandas, for storage and wrangling of data
• scikit-learn (aliased as sklearn), for shallow machine learning models

All of these packages could cover a book on their own, but we will try to give a broad overview of each of them.