Update app.py

app.py

@@ -58,23 +58,44 @@ def get_top_chunks(query): # store a function that gets the most relevant_info a
 client = InferenceClient("google/gemma-3-27b-it")
 
 
-def respond(message, history):
-
-    system_message = "You are a helpful chatbot named scooby, kinda like the cartoon character but not too much like it, who knows alot about pets and their diets and loves helping pet owners, and can only answer questions about pets"
-    messages = [{"role":  "system", "content": system_message}]
+#def respond(message, history):
+#
+ #   system_message = "You are a helpful chatbot named scooby, kinda like the cartoon character but not too much like it, who knows alot about pets and their diets and loves helping pet owners, and can only answer questions about pets"
+  #  messages = [{"role":  "system", "content": system_message}]
 
     
       
-    if history:
-        messages.extend(history)
+   # if history:
+    #    messages.extend(history)
         
-    messages.append({"role": "user", "content": message})
+   # messages.append({"role": "user", "content": message})
+   # response = ""
+   # for messages in client.chat_completion(messages,max_tokens = 2500, stream = True):
+    #    token = messages.choices[0].delta.content
+     #   response += token
+      #  yield response
+
+def respond(message, history):
+    # Get relevant information from knowledge base
+    relevant_chunks = get_top_chunks(message)
+    context = "\n".join(relevant_chunks)
+
+    # Build prompt
+    prompt = (
+        "You are Scooby, a helpful chatbot that only answers questions about pets and their diets.\n"
+        "Use the following relevant information to help answer the user's question:\n\n"
+        f"{context}\n\n"
+        f"User: {message}\nScooby:"
+    )
+
+    # Stream response
     response = ""
-    for messages in client.chat_completion(messages,max_tokens = 2500, stream = True):
-        token = messages.choices[0].delta.content
-        response += token
+    for chunk in client.text_generation(prompt=prompt, max_new_tokens=300, stream=True):
+        response += chunk.token.text
         yield response
 
+
+
 #theme = gr.themes.Origin(primary_hue="orange",secondary_hue="indigo", neutral_hue="teal")