floodlight - A high-level, data-driven sports analyticsframework

floodlight - A high-level, data-driven sports analytics
framework

The increase of available data has had a positive impact on the entire sports domain andespecially sport science (Morgulev et al., 2018). Two major data sources of relevance in this domain are spatiotemporal tracking data of athlete positions as well as manually annotated match event data (Memmert & Raabe, 2018; Stein et al., 2017). These two data types are regularly collected by professional sport organizations in different team invasion games such as football, basketball, or handball (Memmert, 2021). These data sources open up a whole range of new analysis possibilities across multiple (sub)disciplines in the field, including match and performance analysis, exercise physiology, training science, or collective movement behavior analysis. As an example, player tracking data has been used extensively to analyze physical (Castellano et al., 2014) as well as tactical (Rein & Memmert, 2016) performance in football. The floodlight Python package provides a framework to support and automate team sport
data analysis. floodlight is constructed to process spatiotemporal tracking data, event data, and other game meta-information to support scientific performance analyses. floodlight was designed to provide a general yet flexible approach to performance analysis, while simultaneously
providing a user-friendly high-level interface for users with basic programming skills. The package includes routines for most aspects of the data analysis process, including dedicated data classes, file parsing functionality, public dataset APIs, pre-processing routines, common
data models and several standard analysis algorithms previously used in the literature, as well as basic visualization functionality.

@ https://www.instagram.com/prof_daniel_memmert/

Видео floodlight - A high-level, data-driven sports analyticsframework канала Prof. Dr. Daniel Memmert