Training and testing data used in the paper "An equation-of-state-meter of QCD transition from deep learning"

Training and testing data to identify the QCD transition using deep learning and traditional machine learning.<br>1. training_data.csv, testing_iebevishnu.csv, testing_ipglasma.csv<br>There are 723 entries in each row. The first row is the description of the data. The 0th entry in each row is the event id, the entries from 1 to 720 are the pion density distribution at mid-rapidity -- rho(pt, phi) at 15 different pt bins and 48 different azimuthal angle phi bins with phi as the inner loop. The 721st entry is the equation of state type (0 or 1). The 722 entry is extra information of each event.<br>2. training_observables.csv, test_iebe_observables.csv, test_ipglasma_observables.csv <br>There are 87 entries in each row. The first row is the header which describes the data of the following rows. In the following rows, the first entry is the event id, the second entry is the equation of state type (0 or 1), the remaining entries are 85 observables computed from raw spectra, which will be used in various traditional classifiers in machine learning toolbox.<br>The meaning of these entries and the Monte Carlo model used to generate these data can be found in the paper,http://inspirehep.net/record/1503189<br><br><br>