mtry_compare {randomForestVIP}

Mtry Tune via VIPs

Description

A list of data.frames and useful plots for user evaluations of the randomForest hyperparameter mtry.

Usage

mtry_compare(
  formula,
  data = NULL,
  scale = FALSE,
  sqrt = TRUE,
  num_var,
  mvec,
  ...
)

Arguments

formula

an object of class "formula" (or one that can be coerced to that class): a symbolic description of the model to be fitted.

data

an optional data frame containing the variables in the model. By default the variables are taken from the environment which randomForest is called from.

scale

For permutation based measures such as MSE or Accuracy, should the measures be divided by their "standard errors"? Default is False.

sqrt

Boolean value indicating whether importance metrics should be adjusted via a square root transformation. Default is True.

num_var

Optional integer argument for reducing the number of variables to the top 'num_var'. Should be an integer between 1 and the total number of predictor variables in the model or it should be a positive proportion of variables desired.

mvec

Optional vector argument for defining choices of mtry to have the function consider. Should be a vector of integers between 1 and the total number of predictor variables in the model. Or it can be a vector of proportions (strictly less than 1) of the number of predictor variables.

...

Other parameters to pass to the randomForest function.

Value

A list of data.frames, useful plots, and forest objects for user evaluations of the randomForest hyperparameter mtry.

Examples

m <- mtry_compare(factor(Species) ~ ., data = iris, sqrt = TRUE)
m