xgb.importance.interactive: xgboost feature importance interactive table

In Laurae2/Laurae: Advanced High Performance Data Science Toolbox for R

Description

This function interactively plots the table of the feature importance from a xgboost model, provided you give the feature names.

Usage

xgb.importance.interactive(importance, feature_names = NULL, digits = 6,
  pageLength = 20, lengthMenu = c(5, 10, 15, 20, 25, 50, 100, 500),
  gainColor = "lightgreen", coverColor = "lightblue",
  freqColor = "lightgrey")

Arguments

importance	Type: xgboost model, data.table, or data.frame. Your xgboost model to get feature importance from. Or a data.table/data.frame.
feature_names	Type: vector of characters. The feature names to use in the feature importance. If you provide a data.table/data.frame, do not input this argument. Defaults to NULL.
digits	Type: integer. The number of digits to print in the table. Defaults to 6.
pageLength	Type: integer. The number of features to print in a single page. Defaults to 20.
lengthMenu	Type: vector of integers. The selectable number of features to print in a single page. Defaults to c(5, 10, 15, 20, 25, 50, 100, 500).
gainColor	Type: character. The color assigned to bars for Gain depending on its value per feature. Defaults to "lightgreen".
coverColor	Type: character. The color assigned to bars for Cover depending on its value per feature. Defaults to "lightblue".
freqColor	Type: character. The color assigned to bars for Frequency depending on its value per feature. Defaults to "lightgrey".

Value

The "datatable" (not data.table) printed.

Examples

## Not run: 
## my_model <- xgb.train(...)
xgb.importnace.interactive(importance = my_model,
                           feature_names = colnames(data),
                           digits = 6,
                           pageLength = 10)

## End(Not run)

Laurae2/Laurae documentation built on May 8, 2019, 7:59 p.m.