app.py

from huggingface_hub import InferenceClient
import gradio as gr
import logging
import sys

# Set up logging
logging.basicConfig(level=logging.INFO)
logger = logging.getLogger(__name__)

try:
    client = InferenceClient(
        "mistralai/Mixtral-8x7B-Instruct-v0.1"
    )
except Exception as e:
    logger.error(f"Failed to initialize Hugging Face client: {str(e)}")
    sys.exit(1)

def format_prompt(message, history):
    try:
        prompt = "<s>"
        if history:
            for user_prompt, bot_response in history:
                prompt += f"[INST] {user_prompt} [/INST]"
                prompt += f" {bot_response}</s> "
        prompt += f"[INST] {message} [/INST]"
        logger.info(f"Formatted prompt: {prompt}")
        return prompt
    except Exception as e:
        logger.error(f"Error in format_prompt: {str(e)}")
        return None

def generate(message, chat_history, system_prompt, temperature=0.9, max_new_tokens=512, top_p=0.95):
    try:
        logger.info(f"Received message: {message}")
        logger.info(f"System prompt: {system_prompt}")
        
        # Format the full prompt
        if not chat_history:
            full_message = f"{system_prompt}\n\nUser: {message}"
        else:
            full_message = message
            
        formatted_prompt = format_prompt(full_message, chat_history)
        
        if not formatted_prompt:
            return "I encountered an error formatting your message. Please try again."
            
        # Generation parameters
        generate_kwargs = dict(
            temperature=float(temperature),
            max_new_tokens=int(max_new_tokens),
            top_p=float(top_p),
            do_sample=True,
            seed=42,
        )
        
        logger.info("Starting generation with parameters: %s", generate_kwargs)
        
        # Generate response
        response_stream = client.text_generation(
            formatted_prompt,
            **generate_kwargs,
            stream=True,
            details=True,
            return_full_text=False
        )
        
        partial_message = ""
        for response in response_stream:
            if response.token.text:
                partial_message += response.token.text
                yield partial_message

    except Exception as e:
        logger.error(f"Error in generate function: {str(e)}")
        yield f"I encountered an error: {str(e)}"

# Define the default system prompt
DEFAULT_SYSTEM_PROMPT = """You are a supportive AI assistant trained to provide emotional support and general guidance. 
Remember to: 
1. Show empathy and understanding
2. Ask clarifying questions when needed
3. Provide practical coping strategies
4. Encourage professional help when appropriate
5. Maintain boundaries and ethical guidelines"""

# Define the interface
with gr.Blocks() as demo:
    chatbot = gr.Chatbot(height=500)
    msg = gr.Textbox(label="Message", placeholder="Type your message here...")
    
    with gr.Accordion("Advanced Options", open=False):
        system_prompt = gr.Textbox(
            value=DEFAULT_SYSTEM_PROMPT,
            label="System Prompt",
            lines=3
        )
        temperature = gr.Slider(
            minimum=0.1,
            maximum=1.0,
            value=0.9,
            step=0.1,
            label="Temperature"
        )
        max_new_tokens = gr.Slider(
            minimum=64,
            maximum=1024,
            value=512,
            step=64,
            label="Max Tokens"
        )

    clear = gr.Button("Clear")

    def user(user_message, history):
        return "", history + [[user_message, None]]

    def bot(history, system_prompt, temperature, max_new_tokens):
        if not history:
            return history
            
        user_message = history[-1][0]
        history[-1][1] = ""
        
        for chunk in generate(
            user_message, 
            history[:-1], 
            system_prompt, 
            temperature, 
            max_new_tokens
        ):
            history[-1][1] = chunk
            yield history

    msg.submit(
        user,
        [msg, chatbot],
        [msg, chatbot],
        queue=False
    ).then(
        bot,
        [chatbot, system_prompt, temperature, max_new_tokens],
        chatbot
    )

    clear.click(lambda: None, None, chatbot, queue=False)

    gr.Markdown("""
    # PsyAssist - ADVANCING MENTAL HEALTH SUPPORT WITH AI-DRIVEN INTERACTION
    
    **Important Notice**: This is an AI-powered mental health support chatbot. While it can provide emotional support 
    and general guidance, it is not a replacement for professional mental health services. In case of emergency, 
    please contact your local mental health crisis hotline.
    """)

if __name__ == "__main__":
    try:
        demo.queue().launch(show_api=False)
    except Exception as e:
        logger.error(f"Failed to launch Gradio interface: {str(e)}")
        sys.exit(1)