This repository is intended for analyzing STATS Sport VU tracking data and developing the BIMOS (Ball Intercept and Movement for Off-ball Scoring) and BMOS (Ball Movement for Off-Ball Scoring) models for basketball analysis. For a comprehensive explanation of the BIMOS/BMOS models, please refer to our paper on arXiv, which has been accepted for the 11th workshop on Machine Learning and Data Mining for Sports Analytics (MLSA 2024).
Note: This repository is a work in progress. We will continue to update it with additional code for data visualization and further analysis.
You can access the necessary files at the following link: https://drive.google.com/drive/folders/1mJaYJ7xJ95XoBD0LJXhIInSNp_Y_c-pq
Alternatively, you can download the files using the following commands:
curl "https://drive.usercontent.google.com/download?id=1GNaEO4C5kxJuJV2YHw6KpNDbmsAMXPqr&confirm=xxx" -o allevents_dataset.mat
curl "https://drive.usercontent.google.com/download?id=14y-U055lZ-D4snaaLKK1MVwwRc0Y4WS_&confirm=xxx" -o basic_content.zipEnsure that these files are placed in the same directory as the SportVU_ControlField.py file.