Fig. 2
From: DeepDelta: predicting ADMET improvements of molecular derivatives with deep learning
Cross-validation results across benchmark datasets. Correlation plots for Random Forest, ChemProp, DeepDelta, LGBMsub, and Delta LGBMsub following 5 × 10-fold cross-validation. Datasets are sorted by size from smallest (top) to largest (bottom). Coloring is based on data density with the most densely populated regions shown in yellow, least dense regions in blue, and linear interpolation between these groups