Abstract: Time series are ubiquitous in real-world applications, but often add considerable complications to data science workflows. Many machine learning libraries (e.g. scikit-learn) focus on non-temporal data. And even though there are many time series libraries, they are often incompatible with each other.
In this tutorial, we will present sktime - a unified framework for machine learning with time series. sktime covers multiple time series learning problems, including time series transformation, classification and forecasting, among others. In addition, sktime allows you to easily apply an algorithm for one task to solve another (e.g. a scikit-learn regressor to solve a forecasting problem). In the tutorial, you will learn about how you can identify these problems, what their key differences are and how they are related.
To solve these problems, sktime provides various time series algorithms and modular tools for pipelining, ensembling and tuning. In addition, sktime is interfaces with many existing libraries, including scikit-learn, statsmodels and prophet.
You will learn how to use, combine, tune and evaluate different algorithms on real-world data sets. We'll work through all of this step by step using Jupyter Notebooks. Finally, you will find out about how to get involved in sktime's community.
Bio: Franz Kiraly is the founder and a core developer of the open source framework sktime. His research is focused on software engineering for open source and data science, machine learning for structured learning tasks such as time series tasks, and robust empirical and statistical evaluation of algorithms in deployment. Franz held a faculty position at University College London 2013-2020, before he moved to industry R&D in principal data scientist roles.