Fig. 3: Statistical analysis and comparison of three datasets on global minimum adsorption energy (GMAE).
From: A multi-modal transformer for predicting global minimum adsorption energy
a The overview and comparison of three GMAE datasets, with details on dataset size, unique types of surfaces and adsorbates, and brief descriptions of surfaces (S) and adsorbates (A). b Distribution of the GMAE target in three datasets. c The occurrence probability of elements for surface compositions in GMAE datasets. d Distribution of number of atoms of adsorbates in GMAE datasets. e The uniform manifold approximation and projection (UMAP) plot of surface/adsorbate combinations in GMAE datasets, where each combination is depicted by the smooth overlap of atomic positions (SOAP) descriptors of surfaces and RDKit descriptors of adsorbates. This panel illustrates that three GMAE datasets represent distinct chemical spaces, with some areas of overlap.
