Skip to main content

Table 5 Performances of machine learning algorithms for the uncommon solvents in the validation subset

From: Prediction of small-molecule compound solubility in organic solvents by machine learning algorithms

Algorithm Uncommon solvents
MAE MSE R2
PLS 0.5210 0.5638 0.4745
Ridge regression 0.5144 0.5574 0.4805
kNN 0.2780 0.3042 0.7165
DT 0.2792 0.2065 0.8075
ET 0.3128 0.2827 0.7365
RF 0.2568 0.2376 0.7785
SVM 0.2348 0.3139 0.7075
DNN 0.1945 0.1271 0.8815
lightGBM 0.2263 0.1796 0.8326