I was reading scikit-learn documentation for Extremely Randomized Trees and I found this warning:

Warning: Extra-trees should only be used within ensemble methods.

Why is that?


1 Answer 1


In a random forest tree, a random subset of features is available for consideration at each split. Extra-trees takes this a step further by using a random threshold at each split. The idea is that a forest (an ensemble of trees) with a large number of trees that have learned something different from each other (due to the random nature of the features and thresholds at each split) will result in a better prediction than any single tree.

If one were to use a random threshold for each split (what extra-trees do) in a single tree, one would likely end up with a suboptimal tree that predicts poorly.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.