Benedikt Heidrich - Benchmarking Time Series Foundation Models With sktime

sktime is a unified framework for machine learning with time series. This talk presented at PyCon DE PyData Darmstadt 2025 discusses the progress, challenges, and newest features off the press in sktime for working and comparing time series forecasting foundation models with practical examples. 00:00 Video start 00:50 Talk start 01:51 What is sktime? 03:40 Foundation models 07:08 Start of technical difficulties 08:02 End of technical difficulties 09:44 Benchmarking module 11:49 Example: energy forecasting 18:39 Example: hierarchical data forecasting 20:30 Getting started with sktime 21:06 European Summer of Code 22:11 Questions