SAXS nanoparticles for machine learning

This dataset contains real and simulated SAXS data representing nanoparticles samples. The dataset is divided in 4 sub-datasets : synth_xeuss_1800_HR, synth_nano_HR, real_xeuss_1800_HR and real_nano_HR. For synthetic DS, each sample contains the normalized intensity vector I (cm-1), the q vector (A-1), nanoparticles structural parameters used for SaSview simulation, and class label in field 'group'. Real DS are made up of 10 samples each. Each sample is identified by its name (e.g. 'sphere_1') and contains the associated normalized intensity vector and q vector. Intensity vectors are preprocessed with the following steps: 1. buffer background removal 2. time normalization More informations about those data are available in "Automated selection of nanoparticle models for SAXS data analysis using machine learning" (Monge, Deschamps, Amini)