By Robin Brooks
A new book by Jake VanderPlas, eScience Insitute’s director of research in physical sciences, was published in Nov. 2016 by O’Reilly Media. The Python Data Science Handbook is a detailed guide to the most important Python tools for data science, covering IPython, Jupyter, NumPy, Pandas, Matplotlib, Scikit-Learn, and other tools.
Along with its free companion project, A Whirlwind Tour of Python, the book offers a comprehensive introduction to the most important aspects of using Python for gathering, cleaning, processing, modeling, and learning from data across a range of fields.
The content of the book grew out of many of the formal and informal courses and trainings that VanderPlas has taught on these subjects during his three years at the eScience Institute. In keeping with the Institute’s goals of promoting open science and reproducibility, the entire book is available on GitHub in the form of Jupyter notebooks, which are in an open executable document format that researchers can use to share their computational work.