GeneSelectR (version 1.0.1)

plot_metrics: Plot Performance Metrics

Description

This function creates separate plots for f1_mean, recall_mean, precision_mean, and accuracy_mean from a given dataframe, then arranges these plots using cowplot::plot_grid. Requires ggplot2 and cowplot packages.

Usage

plot_metrics(pipeline_results)

Value

A combined ggplot object displaying the performance metrics. - If test metrics are provided, it includes separate plots for F1 mean, recall mean, precision mean, and accuracy mean, along with cross-validation mean score, arranged in a grid layout. - If no test metrics are available, it returns only the cross-validation mean score plot.

Arguments

pipeline_results

An object of class "PipelineResults" containing the performance metrics. This object is expected to contain a dataframe with the columns 'method', 'f1_mean', 'f1_sd', 'recall_mean', 'recall_sd', 'precision_mean', 'precision_sd', 'accuracy_mean', 'accuracy_sd'.

Examples

# Assuming `pipeline_results` is a PipelineResults object with test metrics and CV mean score
pipeline_results <- new("PipelineResults",
                        test_metrics = data.frame(
                        method = c("Method1", "Method2"),
                        f1_mean = c(0.8, 0.85), f1_sd = c(0.05, 0.04),
                        recall_mean = c(0.75, 0.78), recall_sd = c(0.06, 0.05),
                        precision_mean = c(0.85, 0.88), precision_sd = c(0.05, 0.04),
                        accuracy_mean = c(0.9, 0.92), accuracy_sd = c(0.03, 0.02)),
                        cv_mean_score = data.frame(
                        method = c("Method1", "Method2"),
                        mean_score = c(0.88, 0.9), sd_score = c(0.02, 0.02)))

# Plot the performance metrics
metric_plots <- plot_metrics(pipeline_results)
print(metric_plots)