Plot TNA Model Comparison Results — plot_tna

Visualize the distribution of metrics from TNA model comparisons. Supports multiple plot types including histogram (default), boxplot, ridgeline, bar, and density plots.

Usage

plot_tna_comparison(
  results,
  plot_type = "histogram",
  metric_name = "Pearson",
  model_types = NULL,
  binwidth = NULL
)

Arguments

results: A data frame containing comparison results from compare_network_estimation() (the individual element) or run_sampling_analysis().
plot_type: Character string specifying the plot type: "histogram" (default), "boxplot", "ridgeline", "bar", or "density".
metric_name: Character string specifying a single metric to plot. Default: "Pearson". If NULL, all metrics will be plotted.
model_types: Character vector specifying the models to include. If NULL (default), all models in the data will be used.
binwidth: Numeric value for histogram bin width. If NULL (default), calculated automatically.

Value

A ggplot object.

Details

Available plot types:

boxplot: Box plots showing distribution by model type.
ridgeline: Ridge line plots for density comparison.
bar: Bar plots with mean values and error bars.
histogram: Histogram with density overlay (requires metric_name).
density: Density plots with mean/median lines (requires metric_name).

Examples

if (FALSE) { # \dontrun{
library(tna)
data(group_regulation)

# Run comparison
results <- compare_tna_models(
  group_regulation,
  model_types = c("tna", "atna"),
  iterations = 30
)

# Boxplot of Pearson correlation
plot_tna_comparison(results$individual, "boxplot", metric_name = "Pearson")

# Histogram of single metric
plot_tna_comparison(results$individual, "histogram", metric_name = "Pearson")

# All metrics as bar chart
plot_tna_comparison(results$individual, "bar")
} # }