verify_setup.py

#!/usr/bin/env python3
"""
Setup verification script for the Pneumonia Detection ML project.
Run this to check if everything is ready before starting.
"""

import sys
import os
from pathlib import Path

def check_file(filepath, description):
    """Check if a file exists."""
    if Path(filepath).exists():
        size = Path(filepath).stat().st_size
        size_str = f"{size/1024:.1f}KB" if size < 1024*1024 else f"{size/(1024*1024):.1f}MB"
        print(f"  ✓ {description}: {filepath} ({size_str})")
        return True
    else:
        print(f"  ✗ {description}: {filepath} NOT FOUND")
        return False

def check_module(module_name, package_name=None):
    """Check if a Python module can be imported."""
    if package_name is None:
        package_name = module_name
    
    try:
        module = __import__(module_name)
        version = getattr(module, '__version__', 'unknown')
        print(f"  ✓ {package_name}: {version}")
        return True
    except ImportError:
        print(f"  ✗ {package_name}: NOT INSTALLED")
        return False

def main():
    print("="*70)
    print("🔍 PNEUMONIA DETECTION PROJECT - SETUP VERIFICATION")
    print("="*70)
    
    all_good = True
    
    # Check core files
    print("\n📚 Checking Core Project Files...")
    files_to_check = [
        ('pneumonia_detection_ml.ipynb', 'Main Jupyter Notebook'),
        ('download_chest_xray_dataset.py', 'Dataset Download Script'),
        ('train_model.py', 'Training Script'),
        ('predict.py', 'Prediction Script'),
        ('requirements.txt', 'Requirements File'),
        ('README_PNEUMONIA_DETECTION.md', 'Main README'),
        ('QUICK_START_GUIDE.md', 'Quick Start Guide'),
        ('PROJECT_SUMMARY.md', 'Project Summary'),
    ]
    
    for filepath, description in files_to_check:
        if not check_file(filepath, description):
            all_good = False
    
    # Check Python packages
    print("\n📦 Checking Python Packages...")
    packages_to_check = [
        ('numpy', 'NumPy'),
        ('pandas', 'Pandas'),
        ('PIL', 'Pillow'),
        ('cv2', 'OpenCV'),
        ('matplotlib', 'Matplotlib'),
        ('seaborn', 'Seaborn'),
        ('sklearn', 'scikit-learn'),
        ('plotly', 'Plotly'),
        ('jupyter', 'Jupyter'),
    ]
    
    for module_name, package_name in packages_to_check:
        if not check_module(module_name, package_name):
            all_good = False
    
    # Check TensorFlow/Keras separately (special handling)
    print("\n🧠 Checking Deep Learning Frameworks...")
    try:
        import tensorflow as tf
        print(f"  ✓ TensorFlow: {tf.__version__}")
        
        try:
            from tensorflow import keras
            print(f"  ✓ Keras: {keras.__version__}")
        except:
            print(f"  ⚠️  Keras: Available through TensorFlow")
    except ImportError:
        print(f"  ✗ TensorFlow: NOT INSTALLED")
        print(f"     Run: pip install tensorflow")
        all_good = False
    
    # Check dataset
    print("\n📊 Checking Dataset...")
    dataset_dir = Path('./chest_xray')
    if dataset_dir.exists():
        train_dir = dataset_dir / 'train'
        test_dir = dataset_dir / 'test'
        
        if train_dir.exists() and test_dir.exists():
            print(f"  ✓ Dataset found at: {dataset_dir}")
            
            # Count images
            normal_train = len(list((train_dir / 'NORMAL').glob('*.jpeg'))) if (train_dir / 'NORMAL').exists() else 0
            pneumonia_train = len(list((train_dir / 'PNEUMONIA').glob('*.jpeg'))) if (train_dir / 'PNEUMONIA').exists() else 0
            normal_test = len(list((test_dir / 'NORMAL').glob('*.jpeg'))) if (test_dir / 'NORMAL').exists() else 0
            pneumonia_test = len(list((test_dir / 'PNEUMONIA').glob('*.jpeg'))) if (test_dir / 'PNEUMONIA').exists() else 0
            
            print(f"    Training images: {normal_train + pneumonia_train}")
            print(f"      - NORMAL: {normal_train}")
            print(f"      - PNEUMONIA: {pneumonia_train}")
            print(f"    Test images: {normal_test + pneumonia_test}")
            print(f"      - NORMAL: {normal_test}")
            print(f"      - PNEUMONIA: {pneumonia_test}")
        else:
            print(f"  ⚠️  Dataset structure incomplete")
            all_good = False
    else:
        print(f"  ✗ Dataset not found at: {dataset_dir}")
        print(f"     Run: python download_chest_xray_dataset.py")
        all_good = False
    
    # Check for trained model
    print("\n🤖 Checking for Trained Model...")
    model_files = ['pneumonia_model.h5', 'pneumonia_model_best.h5']
    model_found = False
    for model_file in model_files:
        if Path(model_file).exists():
            size = Path(model_file).stat().st_size / (1024*1024)
            print(f"  ✓ Model found: {model_file} ({size:.1f}MB)")
            model_found = True
        else:
            print(f"  ○ Model not yet trained: {model_file}")
    
    if not model_found:
        print(f"     To train: python train_model.py")
        print(f"     Or run the Jupyter notebook")
    
    # Final summary
    print("\n" + "="*70)
    if all_good and model_found:
        print("✅ SETUP COMPLETE - EVERYTHING READY!")
        print("="*70)
        print("\nYou can now:")
        print("  1. Open the notebook: jupyter notebook pneumonia_detection_ml.ipynb")
        print("  2. Make predictions: python predict.py --image xray.jpeg")
        print("  3. Review documentation: README_PNEUMONIA_DETECTION.md")
    elif all_good:
        print("✅ SETUP READY - NEEDS TRAINING")
        print("="*70)
        print("\nNext steps:")
        print("  1. Download dataset: python download_chest_xray_dataset.py")
        print("  2. Train model: python train_model.py")
        print("  3. Or open notebook: jupyter notebook pneumonia_detection_ml.ipynb")
    else:
        print("⚠️  SETUP INCOMPLETE")
        print("="*70)
        print("\nPlease fix the issues above:")
        print("  1. Install missing packages: pip install -r requirements.txt")
        print("  2. Download dataset: python download_chest_xray_dataset.py")
        print("  3. Then run this script again")
    print("="*70)
    
    return 0 if all_good else 1

if __name__ == "__main__":
    sys.exit(main())