Update app.py

app.py

@@ -1,54 +1,64 @@
 import gradio as gr
 from huggingface_hub import InferenceClient
-from collections import defaultdict
 
-# Initialize model client
+"""
+For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
+"""
 client = InferenceClient("Futuresony/future_ai_12_10_2024.gguf")
 
-# Memory storage
-session_histories = defaultdict(list)  # Stores full chat history per session
 
-def format_chat_history(history):
-    """Formats history in a structured way for LLaMA models."""
-    chat_str = ""  
-    for user_msg, bot_response in history:
-        chat_str += f"User: {user_msg}\nAI: {bot_response}\n"
-    return chat_str.strip()  # Remove unnecessary spaces
+def respond(
+    message,
+    history: list[tuple[str, str]],
+    system_message,
+    max_tokens,
+    temperature,
+    top_p,
+):
+    messages = [{"role": "system", "content": system_message}]
 
-def respond(message, history, system_message, max_tokens, temperature, top_p):
-    session_id = id(history)  # Unique session ID
-    session_history = session_histories[session_id]  # Retrieve stored history
+    for val in history:
+        if val[0]:
+            messages.append({"role": "user", "content": val[0]})
+        if val[1]:
+            messages.append({"role": "assistant", "content": val[1]})
 
-    # Add user message to history
-    formatted_history = format_chat_history(session_history)
-    full_input = f"{system_message}\n\n{formatted_history}\nUser: {message}\nAI:"
+    messages.append({"role": "user", "content": message})
 
-    # Generate response
-    response = client.text_generation(
-        full_input,
-        max_new_tokens=max_tokens,
+    response = ""
+
+    for message in client.chat_completion(
+        messages,
+        max_tokens=max_tokens,
+        stream=True,
         temperature=temperature,
         top_p=top_p,
-    )
-
-    # Extract & clean response
-    cleaned_response = response.strip()
+    ):
+        token = message.choices[0].delta.content
 
-    # Update chat history
-    session_history.append((message, cleaned_response))
+        response += token
+        yield response
 
-    return cleaned_response
 
-# Gradio Chat Interface
+"""
+For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
+"""
 demo = gr.ChatInterface(
     respond,
     additional_inputs=[
-        gr.Textbox(value="You are an AI assistant that remembers previous conversations.", label="System message"),
-        gr.Slider(minimum=1, maximum=2048, value=250, step=1, label="Max new tokens"),
+        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 (nucleus sampling)",
+        ),
     ],
 )
 
+
 if __name__ == "__main__":
     demo.launch()