Timeline for what happens to the sample points on pruned leaves
Current License: CC BY-SA 4.0
3 events
when toggle format | what | by | license | comment | |
---|---|---|---|---|---|
Jan 7, 2021 at 3:45 | comment | added | Ben Reiniger♦ |
Quite confident. The trees are built in that fashion, and (post-)pruning essentially undoes some of those splits. Some more reading: scikit-learn.org/stable/auto_examples/tree/… (you could rerun that and add plot_tree , maybe), rdocumentation.org/packages/rpart/versions/4.1-15/topics/…, en.wikipedia.org/wiki/Decision_tree_pruning
|
|
Jan 7, 2021 at 3:18 | comment | added | Student of Statistics | Are you sure about this? You are saying that the decision is on whether or not to split - that is to say, at a node, either all or none leaves are kept. This sounds a bit doubtful intuitively - an unnecessary rigidity. (?) | |
Jan 6, 2021 at 21:01 | history | answered | Ben Reiniger♦ | CC BY-SA 4.0 |