The GitHub repo comes with a reduced dataset (reduced_atlas_dataset.h5), for the purpose of getting familiar with the data. However, one will eventually have to download the full-size data from the CERN website.
All structural code is stored in appropriate .py files. However, for a more comfortable user experience, we provide a Jupyter Notebook where the data will be accessed, the models will be trained, and the main results of this project will be presented.