FEAT: Adding the basic structure of the rfe.py

Author: BrenoFariasdaSilva

rfe.py

@@ -0,0 +1,109 @@
+"""
+================================================================================
+Recursive Feature Elimination (RFE) Automation and Feature Analysis Tool
+================================================================================
+Author      : Breno Farias da Silva
+Created     : 2025-10-07
+Description :
+   This script automates the process of performing Recursive Feature Elimination (RFE)
+   on structured datasets to identify the most relevant features for classification tasks.
+   It provides a fully integrated pipeline — from dataset loading and preprocessing
+   to feature ranking, visualization, and export of analysis reports.
+
+   Core functionalities include:
+      - Dataset validation and safe file handling
+      - Standardization of numeric features using z-score normalization
+      - Recursive Feature Elimination (RFE) with Random Forest as the base estimator
+      - Generation of ranked feature lists with visual and statistical summaries
+      - Boxplot-based visualization of top features by class distribution
+      - Cross-platform sound notification upon completion
+
+Usage:
+   1. Set the `csv_file` variable inside the `main()` function to the dataset path.
+   2. Run the script using:
+      $ make main
+   3. The program will automatically:
+      - Load and clean the dataset
+      - Run RFE to select the most relevant features
+      - Save results and visualizations to the `Feature_Analysis/` directory
+      - Optionally play a notification sound when finished
+
+Output:
+   - Text report (`RFE_results_<Model>.txt`) summarizing feature rankings.
+   - CSV summary of top features with mean and standard deviation per class.
+   - Boxplot visualizations for each selected feature stored in `Feature_Analysis/`.
+
+TODOs:
+   - Add support for additional estimators (e.g., SVM, Gradient Boosting).
+   - Integrate evaluation metrics (F1-score, accuracy, precision, recall, FPR, FNR)
+     directly after feature selection.
+   - Incorporate correlation analysis to remove redundant features.
+   - Extend preprocessing to handle categorical and missing data automatically.
+   - Implement CLI argument parsing for dataset paths and configuration options.
+   - Add parallel RFE runs with different feature subset sizes (1, 2, 5, 10, 15, 20, 25).
+
+Dependencies:
+   - Python >= 3.9
+   - pandas, numpy, seaborn, matplotlib, scikit-learn, colorama
+
+Notes:
+   - The last column of the dataset is assumed to be the target variable.
+   - Only numeric columns are considered for RFE processing.
+   - Sound playback is skipped on Windows platforms by default.
+"""
+
+import atexit # For playing a sound when the program finishes
+import os # For file and directory operations
+import numpy as np # For numerical operations
+import pandas as pd # For data manipulation
+import matplotlib.pyplot as plt # For plotting
+import re # For regular expressions
+import seaborn as sns # For advanced plots
+import platform # For getting the operating system name
+from colorama import Style # For coloring the terminal
+from sklearn.model_selection import train_test_split # For splitting the data
+from sklearn.preprocessing import StandardScaler # For scaling the data (standardization)
+from sklearn.feature_selection import RFE # For Recursive Feature Elimination
+from sklearn.ensemble import RandomForestClassifier # For the Random Forest model
+
+# Macros:
+class BackgroundColors: # Colors for the terminal
+   CYAN = "\033[96m" # Cyan
+   GREEN = "\033[92m" # Green
+   YELLOW = "\033[93m" # Yellow
+   RED = "\033[91m" # Red
+   BOLD = "\033[1m" # Bold
+   UNDERLINE = "\033[4m" # Underline
+   CLEAR_TERMINAL = "\033[H\033[J" # Clear the terminal
+
+# Execution Constants:
+VERBOSE = False # Set to True to output verbose messages
+
+# Sound Constants:
+SOUND_COMMANDS = {"Darwin": "afplay", "Linux": "aplay", "Windows": "start"} # The commands to play a sound for each operating system
+SOUND_FILE = "./.assets/Sounds/NotificationSound.wav" # The path to the sound file
+
+# RUN_FUNCTIONS:
+RUN_FUNCTIONS = {
+   "Play Sound": True, # Set to True to play a sound when the program finishes
+}
+
+# Functions Definitions:
+
+def main():
+   """
+   Main function.
+
+   :return: None
+   """
+   
+   pass
+
+if __name__ == "__main__":
+   """
+   This is the standard boilerplate that calls the main() function.
+
+   :return: None
+   """
+   
+   main() # Call the main function