Computes information about the "proximity" of observations within the YARF model. Given two datasets, information is computed for all pairs of observations. Information returned is the prediction nodes (for all trees) for both objects and their common node in the tree structure plus much information about the nodes.

Usage

compute_raw_proximity_info(
  yarf_mod,
  X1 = NULL,
  X2 = NULL,
  oob_only_1 = TRUE,
  oob_only_2 = TRUE,
  verbose = TRUE
)

Arguments

yarf_mod: A YARF model object.
X1: A n* x p matrix where each row is an observation. If NULL (the default), then this parameter is set to the model's training data.
X2: A n x p matrix where each row is an observation. If NULL (the default), then this parameter is set to the model's training data.
oob_only_1: TRUE (the default) will only return proximity information for trees where each observation is out-of-bag. If X1 is not NULL, this parameter has no effect on output.
oob_only_2: TRUE (the default) will only return proximity information for trees where each observation is out-of-bag. If X2 is not NULL, this parameter has no effect on output.
verbose: Print out periodic messages on progress of the computation of the proximity information. The default is TRUE.

Value

A list indexed by the row number of X1 whose elements are a list indexed by the row number of X2 whose elements are a list indexed by tree number whose elements are a list consisting of the following information: (a) X1 prediction node (b) X2 prediction node, (c) shared node, (d) locations of all three nodes indicating left/right directions, (e) depth differences to shared node where 0 indicates they are in the same node, (f) parent node's cost for all three (where the parent cost of the root is the null model cost)

Author

Adam Kapelner