Pairwise Super-Resolution Dataset with Single electron gun and multi-particle sample

This record contains samples of simulated single electron and multi-particle (electron and photons) samples used in [1] to impact of deep-learning based super-resolution calorimetry for particle reconstruction. The single electron samples were generated with transverse momenta (pT) in [50, 51] GeV, and pseudorapidity (eta) in [-0.01, 0.01] and azimuthal angle (phi) in (-pi, pi) with a 2x downsampling in eta and phi for low resolution. The multi-particle sample contains an electron with pT in [20, 50] GeV, eta in [-2.5, 2.5] and phi in (-pi, pi], the electron is accompanied by randomly sampled 0-3 photons with pT in [5, 25] GeV with a 4x downsampling in eta and phi for low resolution.

All samples were simulated using GEANT4 in the nearly-hermetic 6-layer calorimeter described in [2]. Information about the branches in the TTree format can be found at [3]. In each case, the files contain two separate trees - Low_Tree and High_Tree storing the low resolution and high resolution data respectively.

The train files used are - single_e_train_split{0, 9}.root for the single electron study and multipart_train_250k_split{0, 4}.root for the multi-particle study. The corresponding validation and test files are also provided.

[1] Denoising Graph Super-Resolution towards Improved Collider Event Reconstruction
[2] Configurable Calorimeter Simulation for AI (COCOA)
[3] Output description