crossvalidation_feature_importance: Crossvalidation feature importance

In JimSkinner/toftools: A set of tools for dealing with ToF data

Description

Given a classifier output from 'crossvalidate', return feature importances as either a plot or a data frame. Feature importances are calculated using the caret::varImp function, which computes importances in a different way depending on the classifier. I highly recommend reading the documentation here: https://topepo.github.io/caret/variable-importance.html

Usage

crossvalidation_feature_importance(classifier, n_features = 0,
  override_names = character(0), output_dataframe = FALSE)

Arguments

classifier	Output from 'crossvalidate'
n_features	The number of features to show
override_names	A character vector to override the names of the features. This may be useful if the variable names are too long.
output_dataframe	Default FALSE. If set to TRUE, this function returns a data frame of feature iportances instead of a plot.

Value

Either a plot or a data frame of feature importance.

JimSkinner/toftools documentation built on Oct. 30, 2019, 7:55 p.m.