Fig. 6
From: SIMPD: an algorithm for generating simulated time splits for validating machine learning approaches
(Top): Comparison of the median SA_Score values in the test and training sets for the four different splitting strategies. The plot is divided into two parts for clarity. A: temporal (black crosses) and SIMPD splits (orange circles). B: neighbor (blue squares) and random splits (gray triangles). (Bottom, C: Spatial statistics summary plot \(\sum G\) against \(\sum F'\) for the NIBR medicinal chemistry project data sets with temporal (black crosses), random (gray triangles), neighbor (blue squares), and SIMPD (orange circles) training/test splits