Table 15 Mean square errors on validation and test data of models selected with different scenarios
From: Extended study on atomic featurization in graph neural networks for molecular property prediction
Dataset | Scenario | I | II | III | IV |
|---|---|---|---|---|---|
ESOL (random) | arch. repr. val. \(\downarrow\) test \(\downarrow\) | 2095 F 0.086 0.118 | 2095 \(\underline{\textsc {F}-\textsc {A}}\) 0.083 0.120 | \(\underline{996}\) \(\underline{\textsc {Li}}\) 0.078 0.126 | \(\underline{1743}\) Li 0.078 0.107 |
QM9 | arch. repr. val. \(\downarrow\) test \(\downarrow\) | 914 F 4.531 9.193 | 914 \(\underline{\textsc {Yang}}\) 2.711 14.732 | \(\underline{917}\) Yang 2.659 24.953 | 917 Yang 2.659 24.953 |