ISLP CH8 - Tree-Based Methods

Compléter

(1)

Parties jouées 2

Créer une copie

Introduction to Statistical Learning with Python book

Yukutman

United States

#	Jeux	Précision	temps
1	tachyon 20 Août 2025	% 76	05:12

Voulez-vous apparaître dans le Top 10 de ce jeu? Connectez-vous pour vous identifier.

Créez votre propre jeu gratuite à partir de notre créateur de jeu

Affrontez vos amis pour voir qui obtient le meilleur score dans ce jeu

Créer un défi


Compléter ISLP CH8 - Tree-Based MethodsVersion en ligne Introduction to Statistical Learning with Python book par Yukutman 1 While error is intuitive , it is rarely used for tree construction because it is less sensitive to changes in node class proportions compared to the index and . 2 Recursive binary splitting is a algorithm : it does not reconsider earlier splits once they are made . 3 In cost - complexity pruning , the complexity parameter ? penalizes the number of in the tree , encouraging simpler models . 4 Cross - validation is typically used to select the optimal value of for pruning . 5 In random forests , at each split only a of predictors is considered , which reduces correlation among the trees . 6 The typical number of predictors considered at each split in a random forest is ? p for and p / 3 for . 7 In boosting , each new tree is fit on the of the previous model , and the final prediction is obtained by a weighted of all trees . 8 The three key tuning parameters in boosting are the number of , the rate , and the maximum of each tree . 9 The key advantage of out - of - bag error is that it avoids the need for a separate set .