Skip to main content

Table 1 Cross-validation and testing metrics for the single and ensemble QSPR models trained on the compound solubility dataset

From: Chemically Aware Model Builder (camb): an R package for property and bioactivity modelling of small molecules

Algorithm

\(R^{2}_{CV}\)

RMSECV

\(R^{2}_{0\ test}\)

RMSEtest

A

 GBM

0.90

0.59

0.93

0.52

 RF

0.89

0.62

0.91

0.59

 SVM radial

0.88

0.63

0.91

0.60

B

 Greedy

–

0.57

0.93

0.51

 Linear stacking

0.90

0.57

0.93

0.51

 RF stacking

0.89

0.62

0.92

0.55

  1. The lowest RMSE value on the test set, namely 0.51, was obtained with the greedy and with the linear stacking ensembles.
  2. GBM Gradient Boosting Machine, RF Random Forest, RMSE root mean square error, SVM Support Vector Machine.
Back to article page

Journal of Cheminformatics

ISSN: 1758-2946

Contact us