# Installing the latest version of the transformers library
import os
os.system("pip install ./transformers-4.47.0.dev0-py3-none-any.whl")

# Importing the requirements
import warnings
warnings.filterwarnings("ignore")

import gradio as gr
from src.app.response import caption_image


# Image, text query, and input parameters
image = gr.Image(type="pil", label="Image")
# text = gr.Textbox(label="Question", placeholder="Enter your question here")
max_new_tokens = gr.Slider(
    minimum=20, maximum=160, step=10, value=80, label="Max Tokens"
)

# Output for the interface
answer = gr.Textbox(label="Predicted answer", show_label=True, show_copy_button=True)

# Examples for the interface
examples = [
    [
        "images/cat.jpg",
        80,
    ],
    [
        "images/dog.jpg",
        80,
    ],
    [
        "images/bird.jpg",
        160,
    ],
]

# Title, description, and article for the interface
title = "Visual Question Answering"
description = "Gradio Demo for the PaliGemma 2 Vision Language Understanding and Generation model. This model can answer questions about images in natural language. To use it, upload your image, type a question, select associated parameters, use the default values, click 'Submit', or click one of the examples to load them. You can read more at the links below."
article = "<p style='text-align: center'><a href='https://arxiv.org/abs/2412.03555' target='_blank'>Model Paper</a> | <a href='https://huggingface.co/google/paligemma2-3b-ft-docci-448' target='_blank'>Model Page</a></p>"


# Launch the interface
interface = gr.Interface(
    fn=caption_image,
    inputs=[image, max_new_tokens],
    outputs=answer,
    examples=examples,
    cache_examples=True,
    cache_mode="lazy",
    title=title,
    description=description,
    article=article,
    theme="Nymbo/Nymbo_Theme",
    flagging_mode="never",
)
interface.launch(debug=False)