Skip to main content


Table 1 Comparison of DFT DMs with DMNBO and DMPEOE for the 6703 and 3368 chemical structures of the training and test sets, respectively

From: Machine learning for the prediction of molecular dipole moments obtained by density functional theory

Set Chargea MAE (D) R2/RMSE (D)
Trb NBO 1.44 0.626/1.80
PEOE 0.988 0.650/1.53
Tec NBO 1.44 0.656/1.78
PEOE 0.968 0.659/1.53
  1. aThe DMs were calculated using the DFT geometry optimization
  2. bTraining data set
  3. cTest data set