Dataset: Machine learning-based virtual diagnostics of dielectric laser acceleration

Dataset for the manuscript "Machine learning-based virtual diagnostics of dielectric laser acceleration" The dataset to train the neural network proposed in the manuscript is created by using the tracking code DLAtrack6D [Niedermayer et al., PRAB 20, 111302, 2017] (code is available upon request) and randomly split into training, validation and test set (75% - 15% - 10%). The files training_data_summar.txt and training_data_spectra.txt contain the simulation parameters and the binned spectra (first line: energy bins in eV), respectively. A selection of spectra, also plotted in the manuscript, is given in the files "plot_spectrum_{thetapft}_{arg(e1)}.txt.