Update app.py

app.py

@@ -1,68 +1,44 @@
 import gradio as gr
 from huggingface_hub import InferenceClient
-from web import search  # Import web search tool
+from web import search  # This will now use the `web` tool correctly
 
 client = InferenceClient("Futuresony/future_ai_12_10_2024.gguf")
 
 def is_uncertain(question, response):
     """Check if the model's response is unreliable."""
-    # 1. If response length is too short, it's likely a guess.
-    if len(response.split()) < 4:
+    if len(response.split()) < 4:  # Too short = likely incorrect
         return True
-
-    # 2. If response repeats the question, it might be unsure.
-    if response.lower() in question.lower():
+    if response.lower() in question.lower():  # Repeats question = unsure
         return True
-    
-    # 3. If the response contains generic phrases like "Kulingana na utafiti" (According to research)
     uncertain_phrases = [
         "Kulingana na utafiti", "Inaaminika kuwa", "Ninadhani", 
         "It is believed that", "Some people say", "Inasemekana kuwa"
     ]
     if any(phrase.lower() in response.lower() for phrase in uncertain_phrases):
         return True
-    
     return False
 
 def google_search(query):
-    """Fetch search results from Google."""
-    results = search(query)
+    """Fetch search results using web search."""
+    results = search(query)  # This calls the web tool
     if results:
-        return results[0]  # Return the first result
+        return results[0]  # Return first result
     return "Sorry, I couldn't find an answer on Google."
 
-def respond(
-    message,
-    history: list[tuple[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
-):
+def respond(message, history, system_message, max_tokens, temperature, top_p):
     messages = [{"role": "system", "content": system_message}]
-
     for val in history:
-        if val[0]:
-            messages.append({"role": "user", "content": val[0]})
-        if val[1]:
-            messages.append({"role": "assistant", "content": val[1]})
-
+        if val[0]: messages.append({"role": "user", "content": val[0]})
+        if val[1]: messages.append({"role": "assistant", "content": val[1]})
     messages.append({"role": "user", "content": message})
 
     response = ""
-    
-    for message in client.chat_completion(
-        messages,
-        max_tokens=max_tokens,
-        stream=True,
-        temperature=temperature,
-        top_p=top_p,
-    ):
+    for message in client.chat_completion(messages, max_tokens=max_tokens, stream=True, temperature=temperature, top_p=top_p):
         token = message.choices[0].delta.content
         response += token
-        yield response  # Stream partial responses
-    
-    # If response is unreliable, fetch from Google
+        yield response  # Stream the response
+
+    # If the model's response is unreliable, fetch from Google
     if is_uncertain(message, response):
         google_response = google_search(message)
         yield f"🤖 AI: {response}\n\n🌍 Google: {google_response}"
@@ -73,13 +49,7 @@ demo = gr.ChatInterface(
         gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
         gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
         gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
-        gr.Slider(
-            minimum=0.1,
-            maximum=1.0,
-            value=0.95,
-            step=0.05,
-            label="Top-p (nucleus sampling)",
-        ),
+        gr.Slider(minimum=0.1, maximum=1.0, value=0.95, step=0.05, label="Top-p"),
     ],
 )