Liam Brannigan

Optimise like a scientist

Poorly performing Numpy and Pandas code costs time and money. Try to solve this by just throwing the latest hot new library at it will probably cost you more time and money. In this talk I'll show you instead how to optimise your code like a scientist - that is by building little control experiments, measuring the results and testing, testing, testing to make sure your results remain correct. With this process I will then show you how to throw the hottest new libraries at your problem to make code that runs so fast your computer may just melt.

Target audience: data scientists in particular, but the process & tools for optimising is general to all python users

Mini bio: Liam is the founder of Rho-Signal Data Analytics and provides data science consultancy and training. He builds on his experience developing data science solutions from his time as Lead Data Scientist at Analytics Engines and from his PhD in climate physics at Oxford University. Liam has contributed to a number of open source packages such as xarray and matplotlib. He is a keen communicator on new developments in the python data science world on twitter (https://twitter.com/braaannigan) and now youtube (https://www.youtube.com/channel/UC-J3uR0g7CxCSnx0YFE6R_g)

If you want to submit a talk proposal, please do so here: http://python.ie/meetups

If you wish to sponsor a venue, please contact meetups@python.ie