Dataset: Object condensation: one-stage grid-free multi-object reconstruction in physics detectors, graph, and image data

This dataset is the dataset used to train and test the object condensation particle flow approach described in arxiv:2002.03605. The data can be read with DeepJetCore 3.1 (https://github.com/DL4Jets/DeepJetCore) The entries in the truth array are of dimension (batch, 200, N_truth). The truth inputs are: isElectron, isGamma, isPositron, true_energy, true_x, true_y The entries in the feature array are of dimension (batch, 200, N_features), with the features being: rechit_energy, rechit_x, rechit_y, rechit_z, rechit_layer, rechit_detid The "train.zip" file contains the training sample The "test.zip" file the test sample The main test sample is identical to the training sample in composition, but statistically independent. Other samples can be found in subfolders: test/flatNpart: sample with flat distribution of additional particles in the event w.r.t. each individual particle Test/hiNPart: sample with up to 15 particles per event