Table 4 The mean absolute error, Pearson correlation and average residual in different groups, for prediction of explosive energy (kJ/cc) with leave-one-out CV on the entire dataset using sum over bonds and kernel ridge regression. The groups are sorted by r value rather than MAE since the average explosive energy differs significantly between groups.

From: Applying machine learning techniques to predict the properties of energetic materials

group Ngroup r R2 MAE avg. residual (kJ/cc)
HMX 6 0.968 0.83 0.32 −0.06
Butterfly 10 0.917 0.78 0.18 −0.01
TNT 16 0.854 0.83 0.31 0.10
CL20 6 0.854 0.83 0.22 −0.09
Cubane 12 0.814 0.75 0.58 −0.04
Ring 8 0.548 0.17 0.26 0.20
RDX 6 0.377 0.19 0.28 −0.11
Pyrazole 20 0.254 0.21 0.42 −0.07
Ketone 7 0.099 −0.13 0.25 0.15
Linear 18 0.003 −1.12 0.52 0.00