Table 13 Average test set MSE of the best-performing GCN models trained with the literature representations
From: Extended study on atomic featurization in graph neural networks for molecular property prediction
Representation | Rat \(\downarrow\) | Human \(\downarrow\) | QM9 \(\downarrow\) | ESOL (random) \(\downarrow\) | ESOL (scaffold) \(\downarrow\) |
|---|---|---|---|---|---|
F | 0.182 | 0.218 | 9.193 | 0.118 | 0.166 |
Liu | 0.191 | 0.228 | 80.107 | 0.148 | 0.285 |
Li | 0.223 | 0.223 | 25.155 | 0.107 | 0.201 |
Yang | 0.189 | 0.226 | 24.953 | 0.129 | 0.223 |
Duvenaud | 0.195 | 0.224 | 11.053 | 0.129 | 0.196 |