Update app.py

app.py

@@ -6,182 +6,70 @@ from groq.types.chat.chat_completion_tool_param import ChatCompletionToolParam
 
 # Use the fine-tuned maritime legal model
 MODEL = "nomiChroma3.1"
-client = Groq(api_key=os.environ["GROQ_API_KEY"])
+##client = Groq(api_key=os.environ["GROQ_API_KEY"])
+
+
+from huggingface_hub import InferenceClient
+
+"""
+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("HuggingFaceH4/zephyr-7b-beta")
+
+
+def respond(
+    message,
+    history: list[tuple[str, str]],
+    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]})
+
+    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,
+    ):
+        token = message.choices[0].delta.content
+
+        response += token
+        yield response
+
+"""
+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 a maritime legal assistant with expertise strictly in indian maritime law. Provide detailed legal advice and information based on indian maritime legal principles and regulations.", 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)",
+        ),
+    ],
+    title="Maritime Legal Compliance",
+    description="This chatbot uses the fine tune Llama 3.1 which has the capabilities of responding and helping in legal advices regarding maritime",
+)
 
-# Define your tools if needed (e.g., legal research, document retrieval)
-def legal_tool_function(arguments):
-    # Implement specific legal functions here
-    # Placeholder for legal research or similar functionality
-    return {"result": "Legal tool function response here"}
-
-# Define your tools
-legal_tool: ChatCompletionToolParam = {
-    "type": "function",
-    "function": {
-        "name": "legal_tool_function",
-        "description": "Legal assistant tool: use this for various maritime legal tasks.",
-        "parameters": {
-            "type": "object",
-            "properties": {
-                "arguments": {
-                    "type": "string",
-                    "description": "Arguments for the legal function.",
-                },
-            },
-            "required": ["arguments"],
-        },
-    },
-}
-
-tools = [legal_tool]
-
-def call_function(tool_call, available_functions):
-    function_name = tool_call.function.name
-    if function_name not in available_functions:
-        return {
-            "tool_call_id": tool_call.id,
-            "role": "tool",
-            "content": f"Function {function_name} does not exist.",
-        }
-    function_to_call = available_functions[function_name]
-    function_args = json.loads(tool_call.function.arguments)
-    function_response = function_to_call(**function_args)
-    return {
-        "tool_call_id": tool_call.id,
-        "role": "tool",
-        "name": function_name,
-        "content": json.dumps(function_response),
-    }
-
-def get_model_response(messages, inner_messages, message, system_message):
-    messages_for_model = []
-    for msg in messages:
-        native_messages = msg.get("metadata", {}).get("native_messages", [msg])
-        if isinstance(native_messages, list):
-            messages_for_model.extend(native_messages)
-        else:
-            messages_for_model.append(native_messages)
-
-    messages_for_model.insert(
-        0,
-        {
-            "role": "system",
-            "content": system_message,
-        },
-    )
-    messages_for_model.append(
-        {
-            "role": "user",
-            "content": message,
-        }
-    )
-    messages_for_model.extend(inner_messages)
-
-    try:
-        response = client.chat.completions.create(
-            model=MODEL,
-            messages=messages_for_model,
-            tools=tools,
-            temperature=0.5,
-            top_p=0.65,
-            max_tokens=4096,
-        )
-        return response.choices[0].message
-    except Exception as e:
-        print(f"An error occurred while getting model response: {str(e)}")
-        print(messages_for_model)
-        return None
-
-def respond(message, history, system_message):
-    inner_history = []
-    available_functions = {
-        "legal_tool_function": legal_tool_function,
-    }
-
-    assistant_content = ""
-    assistant_native_message_list = []
-
-    while True:
-        response_message = get_model_response(history, inner_history, message, system_message)
-
-        if response_message is None:
-            return assistant_content, history
-
-        if not response_message.tool_calls and response_message.content is not None:
-            assistant_content += response_message.content
-            assistant_native_message_list.append(response_message)
-            break
-
-        if response_message.tool_calls is not None:
-            assistant_native_message_list.append(response_message)
-            inner_history.append(response_message)
-
-            assistant_content += (
-                "```json\n"
-                + json.dumps(
-                    [
-                        tool_call.model_dump()
-                        for tool_call in response_message.tool_calls
-                    ],
-                    indent=2,
-                )
-                + "\n```\n"
-            )
-            assistant_message = {
-                "role": "assistant",
-                "content": assistant_content,
-                "metadata": {"native_messages": assistant_native_message_list},
-            }
-
-            # Collect responses
-            response_list = [assistant_message]
-
-            for tool_call in response_message.tool_calls:
-                function_response = call_function(tool_call, available_functions)
-                assistant_content += (
-                    "```json\n"
-                    + json.dumps(
-                        {
-                            "name": tool_call.function.name,
-                            "arguments": json.loads(tool_call.function.arguments),
-                            "response": json.loads(function_response["content"]),
-                        },
-                        indent=2,
-                    )
-                    + "\n```\n"
-                )
-                native_tool_message = {
-                    "tool_call_id": tool_call.id,
-                    "role": "tool",
-                    "content": function_response["content"],
-                }
-                assistant_native_message_list.append(
-                    native_tool_message
-                )
-                tool_message = {
-                    "role": "assistant",
-                    "content": assistant_content,
-                    "metadata": {"native_messages": assistant_native_message_list},
-                }
-                response_list.append(tool_message)
-                inner_history.append(native_tool_message)
-
-            return response_list, inner_history
-
-# Update the system prompt to be more relevant to maritime legal assistance
-system_prompt = "You are a maritime legal assistant with expertise in maritime law. Provide detailed legal advice and information based on maritime legal principles and regulations."
-
-# Use gr.Blocks and gr.Chatbot for Gradio 3.x
-with gr.Blocks() as demo:
-    chatbot = gr.Chatbot()
-    system_message_input = gr.Textbox(value=system_prompt, label="System message")
-    message_input = gr.Textbox(label="Message")
-    
-    def process_message(message, history, system_message):
-        responses, updated_history = respond(message, history, system_message)
-        return responses, updated_history
-
-    message_input.submit(process_message, [message_input, chatbot, system_message_input], [chatbot, chatbot])
 
 if __name__ == "__main__":
-    demo.launch()
+    demo.launch()