forest-class: Class '"forest"'

In ParallelForest: Random Forest Classification with Parallel Computing

Description

A forest of decision tree classifiers to be used for ensemble prediction.

Objects from the Class

Objects can be created by calls of the form new("forest", ...).

Slots

n:

Number of observations in dataset used to fit this forest.

p:

Number of independent variables in dataset used to fit this forest.

min_node_obs:

Leaf of any tree in this forest will not be split unless it has more observations than this value.

max_depth:

Maximum depth of any tree in this forest

numsamps:

Number of observations randomly drawn with replacement used to fit a tree in this forest.

numvars:

Number of independent variables randomly drawn without replacement used to fit a tree in this forest.

numboots:

Number of trees in this forest.

numnodes:

Vector with the number of nodes that each tree has in this forest.

flattened.nodes:

Data frame containing information on the nodes of the trees in this forest.

model:

Model frame used to fit this forest.

x:

Design (independent variables) matrix used to fit this forest.

y:

Dependent variable vector used to fit this forest.

fmla:

Formula used to construct the model frame from the data.

depvar.restore.info:

This is a slot that the package needs internally.

ParallelForest documentation built on May 29, 2017, 5:45 p.m.