Phenology/Code/Unsupervised_learning/run_clustering.py

# =============================================================================
# SCRIPT OF EXECUTION - ADVANCED CLUSTERING
# =============================================================================
"""
Helper script to run advanced clustering analysis
with different configurations and datasets.
"""

import subprocess
import sys
import os

def print_menu():
    """Show options menu"""
    print("\n" + "="*80)
    print(" UNSUPERVISED CLUSTERING - NOCCIOLA")
    print("="*80)
    print("\nAvailable options:")
    print("  1. Run full analysis (ResNet50 + K-Means + Hierarchical)")
    print("  2. View previous results")
    print("  3. Install necessary dependencies")
    print("  4. Verify configuration")
    print("  5. Exit")
    print("="*80)

def install_dependencies():
    """Install necessary dependencies"""
    print("\nInstalling dependencies...")
    
    dependencies = [
        'tensorflow',
        'scikit-learn',
        'pandas',
        'numpy',
        'matplotlib',
        'seaborn',
        'tqdm',
        'umap-learn',
        'pillow'
    ]
    
    for dep in dependencies:
        print(f"Installing {dep}...")
        subprocess.run([sys.executable, '-m', 'pip', 'install', dep], check=False)
    
    print("Dependencies installed")

def verify_configuration():
    """Verify that the configuration is correct"""
    print("\nVerifying configuration...")
    
    csv_path = r'C:\Users\sof12\Desktop\ML\Datasets\Carciofo\Robo_GBIF\tags.csv'
    images_dir = r'C:\Users\sof12\Desktop\ML\Datasets\Carciofo\Robo_GBIF'
    
    # Verify CSV
    if os.path.exists(csv_path):
        print(f"CSV found: {csv_path}")
    else:
        print(f"CSV NOT found: {csv_path}")
        return False
    
    # Verify images directory
    if os.path.exists(images_dir):
        # Count images
        img_count = 0
        for root, dirs, files in os.walk(images_dir):
            img_count += sum(1 for f in files if f.lower().endswith(('.jpg', '.jpeg', '.png')))
        print(f"Images directory found: {images_dir}")
        print(f"   Images found: {img_count}")
    else:
        print(f"Images directory NOT found: {images_dir}")
        return False
    
    # Verify TensorFlow
    try:
        import tensorflow as tf
        print(f"TensorFlow installed: {tf.__version__}")
    except ImportError:
        print("TensorFlow NOT installed")
        return False
    
    # Verify GPU
    gpus = tf.config.list_physical_devices('GPU')
    if gpus:
        print(f"GPU available: {len(gpus)} GPU(s)")
    else:
        print("GPU not available, CPU will be used (slower)")
    
    return True

def run_clustering():
    """Run clustering analysis"""
    print("\nRunning clustering analysis...")
    print("This may take several minutes depending on the number of images...")
    
    script_path = os.path.join(
        os.path.dirname(__file__),
        'Clustering_Avanzado.py'
    )
    
    try:
        subprocess.run([sys.executable, script_path], check=True)
        print("\nAnalysis completed successfully!")
        return True
    except subprocess.CalledProcessError as e:
        print(f"\nError during execution: {e}")
        return False

def view_results():
    """View previous results"""
    output_dir = r'C:\Users\sof12\Desktop\ML\Datasets\Nocciola\GBIF\results_clustering_avanzado'
    
    if not os.path.exists(output_dir):
        print("\nNo previous results found. Please run the analysis first.")
        return
    
    print(f"\nResults directory: {output_dir}")
    print("\nFiles found:")
    
    files = os.listdir(output_dir)
    for f in sorted(files):
        size = os.path.getsize(os.path.join(output_dir, f))
        size_mb = size / (1024 * 1024)
        print(f"  - {f} ({size_mb:.2f} MB)")
    
    # Show main CSV if it exists
    csv_path = os.path.join(output_dir, 'results_clustering.csv')
    if os.path.exists(csv_path):
        import pandas as pd
        df = pd.read_csv(csv_path)
        print(f"\nPreview of results_clustering.csv:")
        print(df.head(10))
        print(f"\nTotal rows: {len(df)}")
        print(f"Columns: {list(df.columns)}")
        
        # Show cluster distribution
        if 'cluster_kmeans' in df.columns:
            print(f"\nCluster distribution (K-Means):")
            print(df['cluster_kmeans'].value_counts().sort_index())

def main():
    """Main function"""
    while True:
        print_menu()
        choice = input("\nSelect an option (1-5): ").strip()
        
        if choice == '1':
            if verify_configuration():
                run_clustering()
            else:
                print("\nThere are configuration issues. Please fix them before continuing.")
        
        elif choice == '2':
            view_results()
        
        elif choice == '3':
            install_dependencies()
        
        elif choice == '4':
            verify_configuration()
        
        elif choice == '5':
            print("\nGoodbye!")
            break
        
        else:
            print("\nInvalid option. Please try again.")
        
        input("\nPress Enter to continue...")

if __name__ == '__main__':
    main()