Library: | xclust |
See also: | cartsplit cartsplitopt cartcv leafnum maketr pred prederr prune prunecv prunetot ssr kuva |
Quantlet: | pruneseq | |
Description: | Gives the numbers of leaves and the values of the complexity parameters in the sequence of subtrees, pruned from a tree which has been created by the cartsplit procedure. Prunes using error-complexity criterion. |
Usage: | seq = pruneseq (cs) | |
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. | |
Output: | ||
seq | list of vectors: consists of seq.lnumber, seq.alfa. The elements of seq are vectors with the number of elements equal to the number of trees in the sequence of pruned subtrees of the tree cs. The vector seq.lnumber contains the numbers of leaves in the sequence of the pruned subtrees. The vector seq.alfa contains the values of the complexity parameter alfa. |
; 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 seq=pruneseq(cs) seq
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 Contents of seq.lnumber [1,] 3.000000 [2,] 2.000000 [3,] 1.000000 Contents of seq.alfa [1,] 0.000000 [2,] 60.000000 [3,] 2904.000000
Library: | xclust |
See also: | cartsplit cartsplitopt cartcv leafnum maketr pred prederr prune prunecv prunetot ssr kuva |