plot_feature_importance: SHAP value based Feature Importance plot

In treeshap: Compute SHAP Values for Your Tree-Based Models Using the 'TreeSHAP' Algorithm

SHAP value based Feature Importance plot

Description

This function plots feature importance calculated as means of absolute values of SHAP values of variables (average impact on model output magnitude).

Usage

plot_feature_importance(
  treeshap,
  desc_sorting = TRUE,
  max_vars = ncol(shaps),
  title = "Feature Importance",
  subtitle = NULL
)

Arguments

treeshap	A treeshap object produced with the treeshap function. treeshap.object.
desc_sorting	logical. Should the bars be sorted descending? By default TRUE.
max_vars	maximum number of variables that shall be presented. By default all are presented.
title	the plot's title, by default 'Feature Importance'.
subtitle	the plot's subtitle. By default no subtitle.

Value

a ggplot2 object

See Also

treeshap for calculation of SHAP values

plot_contribution, plot_feature_dependence, plot_interaction

Examples

library(xgboost)
data <- fifa20$data[colnames(fifa20$data) != 'work_rate']
target <- fifa20$target
param <- list(objective = "reg:squarederror", max_depth = 3)
xgb_model <- xgboost::xgboost(as.matrix(data), params = param, label = target,
                              nrounds = 20, verbose = FALSE)
unified_model <- xgboost.unify(xgb_model, as.matrix(data))
shaps <- treeshap(unified_model, as.matrix(head(data, 3)))
plot_feature_importance(shaps, max_vars = 4)

treeshap documentation built on May 29, 2024, 1:27 a.m.