plot_feature_importance {GeneSelectR}

Plot Feature Importance

Description

This function plots the feature importance scores from inbuilt_feature_importance and permutation_importance in the PipelineResults object.

Usage

plot_feature_importance(pipelineresults, top_n_features = 10)

Arguments

pipelineresults

An object of class PipelineResults.

top_n_features

An integer specifying the top N features to plot based on their mean importance.

Value

A list of grid plot objects (ggplot objects) for each feature selection method in the PipelineResults object. Each plot visualizes the top N features based on their mean importance scores, including both inbuilt and permutation importances (if available). The plots are arranged in a grid layout for easy comparison.

Examples

# Assuming `pipelineresults` is a PipelineResults object

pipelineresults <- new("PipelineResults",
                       inbuilt_feature_importance = list("Method1" = data.frame(
                                                         feature = LETTERS[1:10],
                                                         mean_importance = runif(10)),
                                                         "Method2" = data.frame(
                                                         feature = LETTERS[1:10],
                                                         mean_importance = runif(10))),
                       permutation_importance = list("Method1" = data.frame(
                                                                 feature = LETTERS[1:10],
                                                                 mean_importance = runif(10))))

# Plot the feature importance
importance_plots <- plot_feature_importance(pipelineresults, top_n_features = 5)
print(importance_plots)