Figure 3
From: Class imbalance learning with Bayesian optimization applied in drug discovery
Plot of the prediction results by both models. Plot of the 162 molecules with empirically tested antibacterial information and also predicted with top and bottom scores respectively for antibacterial properties by Stokes’ model10. Blue dots represent non-antibacterials; orange dots represent antibacterials. X-axis (Pred_Score_Forest) is the score predicted by our final model, a random forest classifier; Y-axis (Pred_Score_Net) is the score predicted by Stokes’ final model10, a graph neural network.
