Skip to main content

Table 3 Performance of the DRD2 activity model

From: Molecular de-novo design through deep reinforcement learning

Set

Training

Validation

Test

Accuracy

1.00

0.98

0.98

ROC-AUC

1.00

0.99

1.00

Precision

1.00

0.96

0.97

Recall

1.00

0.73

0.82