CeADAR Tech Talk - Up & Running with Polars - Dr Liam Brannigan

Notebook & Dataset:
https://github.com/braaannigan/data-a...
https://github.com/braaannigan/data-a...

The Polars dataframe library has been growing quickly in popularity because it offers an alternative to Pandas that is much faster and scales to large datasets. In this online workshop we will look at why you should consider Polars as a primary data processing tool and how you can get up and running with Polars quickly. We will cover the following topics:

- fundamentals of the Polars API
- running parallel analysis with Polars
- taking advantage of query optimisation
- working with larger-than-memory data
- building a simple time series forecasting pipeline

In addition attendees will be provided with a set of Jupyter notebooks for reference after the workshop.

The workshop will be led by Dr Liam Brannigan. Liam did his PhD in climate physics at the University of Oxford. He went on to become Lead Data Scientist at Belfast-based startup Analytics Engines before founding his own company Rho-Signal. Liam has brought data science projects to production in a number of domains including time series forecasting, NLP and geo-spatial analysis. Liam has been a Polars contributor for over a year with a focus on documentation and accessibility. He has also written many blog posts on effective use of Polars (https://www.rhosignal.com/tags/polars/) and posts frequently on LinkedIn (  / liam-brannigan-9080b214a  )