Numba makes Python 1000x faster!

In this video I introduce the absolute minimum you need to know about Numba which is a just in time compiler for a subset of Python and Numpy. The first half of the video is dedicated to a basic intro and to highlighting a number of very common mistakes people make when using Numba. The remaining video presents a real world-ish simulation problem, shows up to a 1000x acceleration with Numba in both single and multithreaded cases, and concludes with a “reading list” for learning more about Numba.

Channel github: https://github.com/safijari/jack-of-some/projects/1

Find the notebook here: https://gist.github.com/safijari/fa4eba922cea19b3bc6a693fe2a97af7

Deeper topics to discover here
https://numba.pydata.org/numba-doc/dev/user/5minguide.html#other-things-of-interest

Supported python and numpy features
https://numba.pydata.org/numba-doc/dev/reference/pysupported.html
https://numba.pydata.org/numba-doc/dev/reference/numpysupported.html

Important differences from python https://numba.pydata.org/numba-doc/dev/reference/pysemantics.html

Defining types to compile at definition time: http://numba.pydata.org/numba-doc/latest/user/jit.html#eager-compilation

Function factories https://numba.pydata.org/numba-doc/dev/user/generated-jit.html

Experimental version of jitted classes https://numba.pydata.org/numba-doc/dev/user/jitclass.html

Debugging https://numba.pydata.org/numba-doc/dev/user/troubleshoot.html

Dealing with types https://numba.pydata.org/numba-doc/dev/reference/types.html

Ahead of time compilation for deployment https://numba.pydata.org/numba-doc/dev/user/pycc.html

Using approximate fastmath https://numba.pydata.org/numba-doc/dev/user/performance-tips.html#fastmath