Fig. 5
From: Molecular generation by Fast Assembly of (Deep)SMILES fragments
Distribution reproduction experiments’ histograms. Traditional Chinese Medicine database at Taiwan: training set in blue (TCMtrain), generated set in cyan (TCMgene). ChEMBL training set in red (CBLtrain), generated set in orange (CBLgene). The Jaccard index between the training and generated set histograms is shown in the legend (T = x)