The number of trees and tree size in the model controls the processing time for TD_DecisionForest and TD_XGBoost models. When the model size grows more than what can fit in memory, the trees are cached in a local spool space, which may impact the performance of the function compared to the case when all trees fit in memory. Because the SHAP algorithm parses through all branches of a tree and its execution time is heavily impacted by the model size, it is recommended to use it for explainability of a small dataset.