ISLP CH8 - Tree-Based Methods

Completar Frases

(1)

Jugadas 2

Crear copia

Introduction to Statistical Learning with Python book

Yukutman

Estados Unidos

#	Jugador	Acierto	tiempo
1	tachyon 20 de Agosto de 2025	% 76	05:12

¿Quieres aparecer en el Top 10 de este juego? Inicia sesión para identificarte.

Crea tu propio juego gratis desde nuestro creador de juegos

Compite contra tus amigos para ver quien consigue la mejor puntuación en esta actividad

Crear reto


Completar Frases ISLP CH8 - Tree-Based MethodsVersión en línea Introduction to Statistical Learning with Python book por 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 .