Delete app.py

app.py

@@ -1,214 +0,0 @@
-"""
-Gradio web interface for the Medical RAG (Retrieval-Augmented Generation) system.
-
-This application provides a simple interface to ask medical questions and receive
-answers retrieved from a medical knowledge base and augmented with a fine-tuned LLM.
-
-Setup:
-1. Ensure you have the vector database loaded at ./MedQuAD_db
-2. Update the FINE_TUNED_MODEL_ID in model.py to point to your HuggingFace model
-3. Install dependencies: pip install -r requirements.txt
-4. Run: python app.py
-"""
-
-import gradio as gr
-from model import initialize_all, rag_pipeline
-
-
-# Initialize all models on startup
-print("=" * 60)
-print("MEDICAL RAG SYSTEM - Initializing...")
-print("=" * 60)
-try:
-    initialize_all()
-    print("✓ All models loaded successfully!")
-except Exception as e:
-    print(f"✗ Error during initialization: {e}")
-    print("Some models may not be available. The app may still run with reduced functionality.")
-
-print("=" * 60)
-
-
-# ===========================
-# GRADIO INTERFACE FUNCTIONS
-# ===========================
-
-def process_query(user_question, show_context):
-    """
-    Process user question through RAG pipeline.
-    
-    Args:
-        user_question (str): The user's medical question
-        show_context (bool): Whether to show retrieved context details
-        
-    Returns:
-        str: Generated answer with confidence score
-    """
-    if not user_question.strip():
-        return "Please enter a medical question."
-
-    try:
-        answer = rag_pipeline(user_question, top_k=3, detail=show_context)
-        return answer
-    except Exception as e:
-        return f"Error processing query: {str(e)}"
-
-
-def example_questions():
-    """Return example questions for users."""
-    return [
-        "What are the symptoms of type 2 diabetes?",
-        "How is hypertension treated?",
-        "What causes migraines?",
-        "What are the risk factors for heart disease?",
-        "How do I manage chronic pain?"
-    ]
-
-
-# ===========================
-# GRADIO APP CONFIGURATION
-# ===========================
-
-with gr.Blocks(
-    title="Medical Q&A System",
-    theme=gr.themes.Soft(),
-    css="""
-    .header-text {
-        text-align: center;
-        padding: 20px;
-        background: linear-gradient(135deg, #667eea 0%, #764ba2 100%);
-        border-radius: 10px;
-        color: white;
-    }
-    .info-box {
-        background-color: #f0f0f0;
-        border-left: 4px solid #667eea;
-        padding: 15px;
-        margin: 10px 0;
-        border-radius: 5px;
-    }
-    """
-) as demo:
-
-    # Header
-    gr.HTML("""
-    <div class="header-text">
-        <h1>🏥 Medical Question & Answer System</h1>
-        <p>Powered by RAG (Retrieval-Augmented Generation) with Fine-tuned FLAN-T5</p>
-    </div>
-    """)
-
-    # Information box
-    gr.HTML("""
-    <div class="info-box">
-        <strong>⚠️ Disclaimer:</strong> This system is designed for informational purposes only. 
-        It is NOT a substitute for professional medical advice, diagnosis, or treatment. 
-        Always consult with a qualified healthcare provider for medical concerns.
-    </div>
-    """)
-
-    # Main interface
-    with gr.Group():
-        gr.Markdown("### Ask a Medical Question")
-        
-        with gr.Row():
-            with gr.Column(scale=4):
-                question_input = gr.Textbox(
-                    label="Your Question",
-                    placeholder="e.g., What are the symptoms of diabetes?",
-                    lines=3,
-                    info="Enter your medical question in natural language."
-                )
-            
-            with gr.Column(scale=1):
-                show_details = gr.Checkbox(
-                    label="Show Retrieved Context",
-                    value=False,
-                    info="Display source documents"
-                )
-
-        submit_btn = gr.Button(
-            "Get Answer",
-            variant="primary",
-            size="lg"
-        )
-
-        answer_output = gr.Textbox(
-            label="Answer",
-            interactive=False,
-            lines=8,
-            show_copy_button=True
-        )
-
-    # Examples section
-    gr.Markdown("### 📋 Example Questions")
-    
-    examples = example_questions()
-    
-    gr.Examples(
-        examples=[[q] for q in examples],
-        inputs=[question_input],
-        label="Click an example to try it:",
-        run_on_click=False
-    )
-
-    # Information section
-    with gr.Accordion("ℹ️ How This System Works", open=False):
-        gr.Markdown("""
-        This Medical Q&A system uses **Retrieval-Augmented Generation (RAG)**:
-        
-        1. **Retrieval**: Your question is matched against a knowledge base of medical documents
-        2. **Augmentation**: The retrieved documents provide context for generating an accurate answer
-        3. **Generation**: A fine-tuned language model generates an answer based on the retrieved context
-        4. **Confidence**: The system calculates confidence based on how well the retrieved documents match your question
-        
-        **Key Features:**
-        - ✓ Query rewriting for better search
-        - ✓ Semantic search using embeddings
-        - ✓ Context re-ranking for relevance
-        - ✓ Fine-tuned medical model
-        - ✓ Confidence scoring
-        """)
-
-    with gr.Accordion("⚙️ System Configuration", open=False):
-        gr.Markdown("""
-        **Models Used:**
-        - **Embeddings**: sentence-transformers/all-MiniLM-L6-v2
-        - **Query Rewriter**: google/flan-t5-small
-        - **Reranker**: castorini/monot5-base-msmarco
-        - **Generator**: Fine-tuned google/flan-t5-small
-        
-        **Vector Database:**
-        - ChromaDB with persistent storage
-        - ~5000+ medical Q&A chunks from MedQuAD dataset
-        """)
-
-    # Set up interactions
-    submit_btn.click(
-        fn=process_query,
-        inputs=[question_input, show_details],
-        outputs=answer_output,
-        api_name="ask"
-    )
-
-    # Allow Enter key to submit
-    question_input.submit(
-        fn=process_query,
-        inputs=[question_input, show_details],
-        outputs=answer_output
-    )
-
-
-# ===========================
-# LAUNCH CONFIGURATION
-# ===========================
-
-if __name__ == "__main__":
-    demo.launch(
-        server_name="0.0.0.0",
-        server_port=7860,
-        share=False,
-        show_error=True,
-        show_tips=True,
-        debug=False
-    )