Numba

The Numba Python compiler from continuum analytics helps make highly parallelizable, incredibly powerful performance from an interpreted language a reality.

Note

Note: The documentation on the official pydata website provides a comprehensive overview of what Numba is and how you can leverage it in your own Python programs. You can find it at http://numba.pydata.org/#.

In this section, we'll have a look at the ecosystem surrounding Numba, which takes its form in the shape of Anaconda. We'll also look at how you can then leverage Numba alongside numerous other packages in order to effectively and efficiently perform analysis of big data. We'll cover some of the basics of Numba and then work our way into the more complex aspects such as utilizing GPUs and APUs within our program.

Overview

Numba is very cool in the sense that it generates optimized machine code from pure Python code using the LLVM compiler infrastructure. By making slight modifications to our existing code, we can see incredible...