Visualization/Plotting of Results by using The RStudio

GAN Metrics Analysis in R

Overview

This project provides a comprehensive analysis and visualization of metrics obtained during GAN training

It includes Dice Score and Loss metrics, plotted as line plots, bar charts, and boxplots, as well as multi-panel figures suitable for publication...

Folder Structure

• data/raw/ : Original Excel files with metrics.
• data/processed/ : Processed CSV files generated by the analysis.
• Scr : R scripts containing all code.
• outputs/ : Generated figures and plots.

Features

• Dice Score evolution across epochs (training vs validation)
• Loss component trends (generator & discriminator)
• Bar charts for final epoch comparison
• Boxplots showing distributions across epochs
• Combined multi-panel figure for publication

Requirements

• R >= 4.2
• Packages: tidyverse, readxl, patchwork

Usage

1. Place your raw Excel data in data/raw/.
2. Run the analysis script:

source("scripts/analysis.R")
results <- main()