Journal of Cheminformatics

Table 5 Prediction of the DFT DM by random forests using a combination of DM_NBO and DM_PEOE with different type of descriptors

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

Models	Training set^a		Test set
Models	MAE (D)	R²/RMSE (D)	MAE(D)	R²/RMSE (D)
MACCS^b	0.460	0.853/0.707	0.444	0.872/0.680
RDF^b,c	0.533	0.817/0.791	0.522	0.839/0.767
PchmDM^b,d	0.540	0.819/0.786	0.533	0.839/0.765
CDK^b	0.538	0.815/0.793	0.525	0.837/0.765

^aOOB estimation
^bCombining DM_PEOE, and DM_NBO
^cRDF pairs NBO charges
^dPchmDM NBO charges

Back to article page

ISSN: 1758-2946

Contact us

Submission enquiries: journalsubmissions@springernature.com