Is there a way to get the instances used for class

问题:

The getTree function in randomForest package in R displays the structure of the a particular tree used in the random forest.

Here is an example on the iris dataset

library(randomForest)
data(iris)
rf <- randomForest(Species ~ ., iris)
getTree(rf, 1)

This shows the output of tree #1 of 500:

   left daughter right daughter split var split point status prediction
1              2              3         3        2.50      1          0
2              0              0         0        0.00     -1          1
3              4              5         4        1.65      1          0
4              6              7         4        1.35      1          0
5              8              9         3        4.85      1          0
6              0              0         0        0.00     -1          2
7             10             11         2        3.10      1          0
8             12             13         4        1.55      1          0
9              0              0         0        0.00     -1          3
10             0              0         0        0.00     -1          3
11             0              0         0        0.00     -1          2
12            14             15         2        2.55      1          0
13             0              0         0        0.00     -1          2
14            16             17         2        2.35      1          0
15             0              0         0        0.00     -1          3
16             0              0         0        0.00     -1          3
17             0              0         0        0.00     -1          2

The leaves are the nodes with 0 leaf daughter and 0 right daughter.

Is there a way I can get which instances (rows of the iris data set) are in those leaves?
Like Node 2 which is a leaf has instance 2,3,4 from iris dataset all classified as 1.

Any help will be much appreciated.

回答1:

Building on this answer:

Extract a subset of tree from random forest model for prediction

There may be an access function I'm not aware of, but the following manual approach seems to work:

rf <- randomForest(Species ~ ., iris, keep.forest=TRUE)
pr_nodes <- attr(predict(rf, iris, nodes=TRUE),'nodes')

pr_nodes[,1] # array of 150 node assignments for 1st of 500 trees