Speeding up the Python Lectures
5 Dec 2018
QuantEcon has been working to speed up the Python lectures with the help of just-in-time compilation from Numba. Just-in-time (jit) compilation allows code to be compiled on the fly - a flexible combination of ahead-of-time compilation and interpretation (standard Python is interpreted).
By jitting the lectures, we’ve seen speed improvements of about 1 to 2 orders of magnitude. A discussion of the benefits of Numba with Python is in No, Python is Not Too Slow for Computational Economics.
The majority of jitted lectures use fixed point iteration to find the model’s solution, some examples include The Stochastic Optimal Growth Model, The Lucas Asset Model and Search With Learning.
Thanks to Natasha Watkins for work on this improvement.
Published by: Natasha Watkins