| Library: | xclust |
| See also: | cartsplit cartsplitopt cartcv maketr pred prederr prune prunecv pruneseq prunetot ssr kuva |
| Quantlet: | leafnum | |
| Description: | Gives the number of leaves (terminal nodes) in a regression tree. |
| Usage: | ln = leafnum(cs, node) | |
| Input: | ||
| cs | list of vectors: data structure which represents a binary tree and is produced by cartsplit procedure, contains vectors cs.val, cs.vec, cs.mean, cs.ssr, cs.nelem. See cartsplit for the description of cs. | |
| node | integer >= 1: gives the index of the root node of the subtree whose number of leaves will be calculated. If node=1, the number of leaves in the hole tree is calculated, if node=2, the number of leaves in the left subtree is calculated, if node=3, the number of leaves in the left subtree of the left subtree (if it exists) is calculated. See cartsplit for the explanation how the tree is represented as a vector. | |
| Output: | ||
| ln | integer >= 1: number of leaves in the specified subtree. | |
; loads the library xclust
library ("xclust")
;let us generate a tree by cartsplit procedure
x1=#(0,0,0,0,1,1,1,1,1,2)
x2=#(0,0,0,0,0,0,0,1,1,1)
x=x1~x2
y=#(0,0,0,0,100,100,100,120,120,120)
cs=cartsplit(x,y,#(0,1))
cs
ln1=leafnum(cs,1)
ln1
ln2=leafnum(cs,2)
ln2
ln3=leafnum(cs,3)
ln3
Content of object cs.val.split0 [1,] 0 [2,] 1,2 Content of object cs.val.split1 [1,] NaN Content of object cs.val.split2 [1,] 0 Content of object cs.val.split3 [1,] NaN Content of object cs.val.split4 [1,] NaN Content of object cs.vec [1,] 1.000000 [2,] NaN [3,] 2.000000 [4,] NaN [5,] NaN Content of object cs.mean [1,] 66.000000 [2,] 0.000000 [3,] 110.000000 [4,] 100.000000 [5,] 120.000000 Content of object cs.var [1,] 29640.000000 [2,] 0.000000 [3,] 600.000000 [4,] 0.000000 [5,] 0.000000 Content of object cs.nelem [1,] 10.000000 [2,] 4.000000 [3,] 6.000000 [4,] 3.000000 [5,] 3.000000 Content of object ln1 [1,] 3.000000 Content of object ln2 [1,] 1.000000 Content of object ln3 [1,] 2.000000
| Library: | xclust |
| See also: | cartsplit cartsplitopt cartcv maketr pred prederr prune prunecv pruneseq prunetot ssr kuva |