EIC Pythia Events

Dataset used in the publication Point cloud-based diffusion models for the Electron-Ion Collider, arXiv: 2410.2242The dataset consists of DIS events for the EIC generated by Pythia8  at a representative CM energy for electron-nucleus collisions at the EIC sqrt(s) = 105 GeV. We avoid the low-Q^2 photoproduction region by imposing a cut of Q^2>25 GeV^2. We include all particles in the rapidity range |y|<5, and we do not impose a lower cut on the transverse momentum. 

In the h5 files we include the following entries:['data'] = [N,50, 13] the set of input particles with features : [eta_p + eta_h, phi_p + phi_h, log(h_pT/e_pT), charge,is_proton,is_neutron, is_kaon, is_charged_pion, is_neutrino,is_muon, is_electron,is_photon,]The train_eic.h5 and val_eic.h5 are used for training and validation while results are compared to the test_eic.h5 file