Update app.py

app.py

@@ -1,66 +1,113 @@
 import gradio as gr
 from huggingface_hub import InferenceClient
 
+# Import the 'gr.OAuthToken' type for Gradio to handle the OAuth token automatically.
+
+# It is a best practice to define the model ID separately for clarity.
+MODEL_ID = "E5K7/eshalskoibito"
+
 
 def respond(
     message,
     history: list[dict[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
+    system_message: str,
+    max_tokens: int,
+    temperature: float,
+    top_p: float,
     hf_token: gr.OAuthToken,
 ):
     """
-    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
+    Handles the chatbot's response by sending a request to the Hugging Face Inference API.
+
+    Args:
+        message (str): The user's message.
+        history (list): The list of previous conversation turns.
+        system_message (str): The system message for the model.
+        max_tokens (int): The maximum number of new tokens to generate.
+        temperature (float): The sampling temperature.
+        top_p (float): The top-p value for nucleus sampling.
+        hf_token (gr.OAuthToken): The Hugging Face OAuth token for authentication.
+
+    Yields:
+        str: The generated response, streamed token by token.
     """
-    client = InferenceClient(token=hf_token.token, model="E5K7/eshalskoibito")
+    # Ensure the Hugging Face token is available before proceeding.
+    if hf_token is None:
+        raise gr.Error("You must log in to use the chatbot!")
+
+    # Initialize the InferenceClient with the provided token.
+    client = InferenceClient(token=hf_token.token, model=MODEL_ID)
 
     messages = [{"role": "system", "content": system_message}]
 
-    messages.extend(history)
+    # Format the chat history for the client, which expects a list of dictionaries.
+    messages.extend(
+        [
+            {"role": turn["role"], "content": turn["content"]}
+            for turn in history
+        ]
+    )
 
     messages.append({"role": "user", "content": message})
 
     response = ""
-
-    for message in client.chat_completion(
-        messages,
+    for token in client.chat_completion(
+        messages=messages,
         max_tokens=max_tokens,
         stream=True,
         temperature=temperature,
         top_p=top_p,
     ):
-        choices = message.choices
-        token = ""
-        if len(choices) and choices[0].delta.content:
-            token = choices[0].delta.content
-
-        response += token
-        yield response
+        if token.choices and token.choices[0].delta.content:
+            response += token.choices[0].delta.content
+            yield response
 
 
-"""
-For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
-"""
+# Create the ChatInterface with updated parameters for better user experience.
 chatbot = gr.ChatInterface(
     respond,
     type="messages",
+    # Add a title and description for better context.
+    title="Eshalskoibito Chatbot",
+    description=f"Interact with the model: **{MODEL_ID}**",
     additional_inputs=[
-        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.Textbox(
+            value="You are a friendly Chatbot.",
+            label="System message",
+            info="Define the persona and behavior of the chatbot.",
+        ),
+        gr.Slider(
+            minimum=1,
+            maximum=2048,
+            value=512,
+            step=1,
+            label="Max new tokens",
+            info="The maximum number of tokens to generate in the response.",
+        ),
+        gr.Slider(
+            minimum=0.1,
+            maximum=4.0,
+            value=0.7,
+            step=0.1,
+            label="Temperature",
+            info="Controls the randomness of the output. Higher values lead to more creative responses.",
+        ),
         gr.Slider(
             minimum=0.1,
             maximum=1.0,
             value=0.95,
             step=0.05,
             label="Top-p (nucleus sampling)",
+            info="Filters out low-probability tokens. Lower values make the response more focused.",
         ),
     ],
+    # Add a parameter to save chat history locally in the user's browser.
+    # This prevents conversation mixing between multiple users.
+    save_history=True,
 )
 
 with gr.Blocks() as demo:
+    # Use gr.LoginButton() and pass the oauth token to the chatbot function.
     with gr.Sidebar():
         gr.LoginButton()
     chatbot.render()