GitHub Repo for this session: https://github.com/ltsaprounis/dsf-ts-forecasting
Join us to experience the end-to-end workflow for a real-world time series forecasting project using python. In this hands-on workshop we’ll take a real-life dataset containing over 50 time series for Influenza like Illness (ILI) incidence in the US and look at how to produce the most accurate forecasts. We’ll use statistical methods such as exponential smoothing and machine learning methods (e.g. lightgbm) and make them compete for forecasting accuracy. We’ll also uncover the important principles and learnings of a forecasting project that include the metrics to select for optimal KPIs and how to best evaluate your models.
The workshop is inspired by a real AI use case at GSK Consumer Healthcare where we are committed to delivering everyday health and humanity to over a billion people around the world.
A basic understanding of Python and Machine Learning is recommended for this workshop.