fw.xpy: Forward variable selection for PDP explanation

In g-rho/eXplaAInability: Numerical and visual tools to analyse the explainability of partial dependence functions

Description

Computes forward variable selection for partial dependence function based on explainability.

Usage

fw.xpy(model, x, target, ...)

Arguments

model	A model with corresponding predict function that returns numeric values.
x	Data frame.
target	Character specifying the name of the (numeric) target variable (must be contained in data).
...	Further arguments to be passed to the predict() method of the model.

Value

Object of class vsexp containing the following elements:

selection.order	Vector of variable names in order of their entrance to the PD function during the variable selection process.
explainability	Explainabilities after the different iterations of the variable selection.
details	A data frame containing the explainabilities for all variables (rows) if they were entered in the model at each step (columns).

Author(s)

[e-mail]

References

• Szepannek, G. (2019): How Much Can We See? A Note on Quantifying Explainability of Machine Learning Models, arXiv:1910.13376 [stat.ML].

Examples

## Not run: 
library(pdp)
library(randomForest)
data(boston)
set.seed(42)
boston.rf <- randomForest(cmedv ~ ., data = boston)
vs <- fw.xpy(boston.rf, boston, "cmedv")
vs

plot(vs)

## End(Not run)

g-rho/eXplaAInability documentation built on April 6, 2021, 1:42 a.m.