app.py

import gradio as gr
import os
from multimodal_ai import MultimodalAI

examples = [
    {"text": "What is this image?", "files": ["sample/ducky.jpg"]},
    {"text": "Can you summarize and translate this?", "files": ["sample/sermon.mp4"]}
]

# Function to print the like/dislike data
def print_like_dislike(x: gr.LikeData):
    print(x.index, x.value, x.liked)

# Function to add a message to the chat history
def add_message(history, message):
    for x in message["files"]:
        history.append(((x,), None))
    if message["text"] is not None:
        history.append((message["text"], None))
    return history, gr.MultimodalTextbox(value=None, interactive=False), message["text"]

# Function to generate the bot's response
def bot(history, user_input):
    response = ai_assistant.generate_response(user_input)
    if history:
        history[-1][1] = response
    return history

def main():
    # Define the Gradio interface using Blocks
    with gr.Blocks(fill_height=True) as webApp:
        chatbot = gr.Chatbot(
            elem_id="chatbot",
            bubble_full_width=False,
            scale=1,
        )
        with gr.Column(scale=8):
            chat_input = gr.MultimodalTextbox(interactive=True,
                                              #file_count="multiple",
                                              placeholder="Enter message or upload file...", 
                                              show_label=False)
            user_input = gr.State()
            chat_msg = chat_input.submit(add_message, [chatbot, chat_input], [chatbot, chat_input, user_input])
            bot_msg = chat_msg.then(bot, [chatbot, user_input], chatbot, api_name="bot_response")
            bot_msg.then(lambda: gr.MultimodalTextbox(interactive=True), None, [chat_input])
            chatbot.like(print_like_dislike, None, None)

        with gr.Column(scale=3):
            # Add examples
            gr.Examples(
                examples=examples,
                inputs=chat_input,
                label="Click on an example to try it"
            )
        

    # Start the queue and launch the interface
    webApp.queue()
    webApp.launch(share=True)

if __name__ == "__main__":
    # Initialize llm
    ai_assistant = MultimodalAI()
    main()