Decision trees (the base learner in random forests and gradient boosted decision trees) are effectively nested “if” statements, and predictions from tree ensemb

Decision trees (the base learner in random forests and gradient boosted decision trees) are effectively nested “if” statements, and predictions from tree ensembles (RF and GBDT) are effectively averages of predictions from those nested “if” statements.