app.py

import gradio as gr
from transformers import pipeline
from reportlab.lib.pagesizes import letter
from reportlab.pdfgen import canvas

# Load a pre-trained image classification model from Hugging Face
# Replace with a fine-tuned model if available
home_inspector = pipeline("image-classification", model="google/vit-base-patch16-224")

# Function to analyze the uploaded image
def inspect_home(image):
    # Analyze the image using the Hugging Face model
    results = home_inspector(image)
    
    # Format the results
    detected_issues = {result["label"]: round(result["score"], 2) for result in results}
    
    # Generate repair suggestions based on detected issues
    suggestions = []
    for issue in detected_issues:
        suggestion = f"For {issue}, consider inspecting the area closely and consulting a professional if necessary."
        suggestions.append(suggestion)
    
    # Generate a PDF report
    generate_report(detected_issues, suggestions)
    
    # Return results and suggestions
    return detected_issues, suggestions

# Function to generate a PDF report
def generate_report(issues, suggestions, filename="home_inspection_report.pdf"):
    c = canvas.Canvas(filename, pagesize=letter)
    
    # Add a title
    c.setFont("Helvetica-Bold", 16)
    c.drawString(100, 750, "Home Inspection Report")
    
    # Add a subtitle
    c.setFont("Helvetica", 12)
    c.drawString(100, 730, "AI-Powered Home Inspector")
    
    # Add issues and suggestions
    y = 700
    for issue, suggestion in zip(issues, suggestions):
        c.setFont("Helvetica-Bold", 12)
        c.drawString(100, y, f"Issue: {issue} (Confidence: {issues[issue]})")
        c.setFont("Helvetica", 10)
        c.drawString(100, y - 20, f"Suggestion: {suggestion}")
        y -= 40
    
    # Save the PDF
    c.save()

# Custom CSS for a modern blue and silver interface
custom_css = """
body {
    background: linear-gradient(135deg, #1e3c72, #2a5298);
    color: white;
    font-family: 'Helvetica', sans-serif;
}
.gradio-container {
    background: rgba(255, 255, 255, 0.1);
    border-radius: 10px;
    padding: 20px;
    box-shadow: 0 4px 6px rgba(0, 0, 0, 0.1);
}
.gradio-input, .gradio-output {
    background: rgba(255, 255, 255, 0.2);
    border: 1px solid rgba(255, 255, 255, 0.3);
    border-radius: 5px;
    padding: 10px;
    color: white;
}
.gradio-button {
    background: #4a90e2;
    color: white;
    border: none;
    border-radius: 5px;
    padding: 10px 20px;
    font-size: 16px;
    cursor: pointer;
    transition: background 0.3s ease;
}
.gradio-button:hover {
    background: #357abd;
}
"""

# Gradio interface
with gr.Blocks(css=custom_css) as demo:
    gr.Markdown("# 🏠 AI-Powered Home Inspector")
    gr.Markdown("Upload a photo of your home, and the AI will detect issues and provide repair suggestions.")
    
    with gr.Row():
        with gr.Column():
            image_input = gr.Image(label="Upload a Photo", type="pil")
            inspect_button = gr.Button("Inspect Home", variant="primary")
        
        with gr.Column():
            issues_output = gr.Label(label="Detected Issues")
            suggestions_output = gr.Textbox(label="Repair Suggestions", lines=5)
    
    # Link the button to the function
    inspect_button.click(
        fn=inspect_home,
        inputs=image_input,
        outputs=[issues_output, suggestions_output]
    )
    
    gr.Markdown("### Report Generated")
    gr.Markdown("A very detailed PDF report has been saved as `home_inspection_report.pdf`.")

# Launch the app
demo.launch()