This figure shows the precision-recall curves (PRCs) for each supervised learning model trained with Input A and tested on the external dataset.

(A), (B), (C), (D) and (E) display the PRCs for the logistic regression (LR), support vector machine (SVM), decision tree (DT), random forest (RF), and gradient-boosted decision tree (GBDT) models, respectively. For each model, the graphs evaluated using a One vs. Rest (OvR) approach are shown on the top and a One vs. One (OvO) approach are shown on the bottom (note that only the OvO PRCs for consecutive phases are shown). See S2 Fig for more details on their interpretation. (PDF)