Fig. 4From: An exploration strategy improves the diversity of de novo ligands using deep reinforcement learning: a case for the adenosine A2A receptorPerformance of five different machine learning models based on fivefold cross validation in the A2AR set with different metrics, including AUC of ROC curve (a), MCC, Sensitivity, Specificity and Accuracy values (b). Except for specificity, the RF achieved highest scores among these models based on such measurementsBack to article page