Skip to main content

Table 4 The results of three different blending methods

From: Comparison and improvement of the predictability and interpretability with ensemble learning models in QSPR applications

 

Training dataset

Test dataset

\(R^{2}\)

RMSE (nm)

\(R_{pred}^{2}\)

RMSE (nm)

Uniform blending

0.988

13.26

0.921

31.35

Linear blending

0.992

10.25

0.922

31.05

Any blending

0.996

7.84

0.931

29.11