Table 3 Comparisons of performance with state-of-the-art methods on regression datasets, splitting the datasets by random splitting in a ratio of 8:1:1 (lower values are better)
From: Double-head transformer neural network for molecular property prediction
Methods | Lipophilicity | PDBbind |
|---|---|---|
MolNet [1] | \(0.655\pm 0.036\) | \(1.920\pm 0.070\) |
RF on Morgan [19] | \(0.823\pm 0.035\) | \(2.083\pm 0.324\) |
FFN on Morgan [19] | \(0.928\pm 0.044\) | \(2.778\pm 0.599\) |
FFN on Morgan counts [19] | \(0.874\pm 0.043\) | \(2.901\pm 0.812\) |
FFN on RDKit [19] | \(0.735\pm 0.039\) | \(2.020\pm 0.376\) |
DMPNN [19] | \(0.582\pm 0.024\) | \(1.945\pm 0.298\) |
Ours | \({{ {0.577}}}\pm {{ {0.049}}}\) | \({{ {1.771}}}\pm {{ {0.300}}}\) |