Update app.py

app.py

@@ -45,14 +45,7 @@ def retrieve_docs(query, k=5):
     return retrieved_docs
 
 # RAG pipeline integrated into respond function
-def respond(
-    message,
-    history: list[tuple[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,  # Keeping top_p as an input, though Gemini doesn’t use it directly
-):
+def respond(message, system_message, max_tokens, temperature):
     # Preprocess the user message
     preprocessed_query = preprocess_text(message)
     
@@ -60,17 +53,13 @@ def respond(
     retrieved_docs = retrieve_docs(preprocessed_query, k=5)
     context = "\n".join(retrieved_docs['text'].tolist())
     
-    # Construct the prompt with system message, history, and RAG context
+    # Construct the prompt with system message and RAG context, asking for structured response
     prompt = f"{system_message}\n\n"
-    for user_msg, assistant_msg in history:
-        if user_msg:
-            prompt += f"User: {user_msg}\n"
-        if assistant_msg:
-            prompt += f"Assistant: {assistant_msg}\n"
     prompt += (
         f"Query: {message}\n"
         f"Relevant Context: {context}\n"
-        f"Generate a short, concise, and to-the-point response to the query based only on the provided context."
+        f"Generate a short, concise response to the query based only on the provided context. "
+        f"Format the response as a structured list (e.g., bullet points or numbered items) instead of a paragraph."
     )
     
     # Generate response with Gemini
@@ -88,37 +77,27 @@ def respond(
             answer = answer[:last_period + 1]
         else:
             answer += "."
+    
+    return answer
 
-    # Format the output with Gradio markdown for better readability
-    formatted_answer = f"""
-    <div style='background-color:#f0f0f0; padding: 10px; border-radius: 5px;'>
-        <h3 style='color:#333; font-weight:bold;'>Assistant's Response:</h3>
-        <p style='color:#555;'>{answer}</p>
-    </div>
-    """
-    # Yield the formatted response
-    yield formatted_answer
+# Simple Gradio Interface
+def chatbot_interface(message, system_message, max_tokens, temperature):
+    return respond(message, system_message, max_tokens, temperature)
 
-# Gradio Chat Interface
-demo = gr.ChatInterface(
-    respond,
-    additional_inputs=[
+demo = gr.Interface(
+    fn=chatbot_interface,
+    inputs=[
+        gr.Textbox(label="Your Query", placeholder="Enter your medical question here..."),
         gr.Textbox(
             value="You are a medical AI assistant diagnosing patients based on their query, using relevant context from past records of other patients.",
-            label="System message"
+            label="System Message"
         ),
-        gr.Slider(minimum=1, maximum=2048, value=150, step=1, label="Max new tokens"),
+        gr.Slider(minimum=1, maximum=2048, value=150, step=1, label="Max Tokens"),
         gr.Slider(minimum=0.1, maximum=4.0, value=0.75, step=0.1, label="Temperature"),
-        gr.Slider(
-            minimum=0.1,
-            maximum=1.0,
-            value=0.95,
-            step=0.05,
-            label="Top-p (nucleus sampling)",  # Included but not used by Gemini
-        ),
     ],
+    outputs=gr.Textbox(label="Response"),
     title="🏥 Medical Chat Assistant",
-    description="A chat-based medical assistant that diagnoses patient queries using AI and past records."
+    description="A simple medical assistant that diagnoses patient queries using AI and past records, providing structured responses."
 )
 
 if __name__ == "__main__":