app.py

import gradio as gr
import fitz  # PyMuPDF
from PIL import Image

def analysis_process(pdf_file):
    # Open the PDF file
    document = fitz.open(pdf_file.name)
    # Get the first page
    page = document.load_page(0)
    # Render the page to a pixmap
    pix = page.get_pixmap()
    # Save the pixmap as an image
    image = Image.frombytes("RGB", [pix.width, pix.height], pix.samples)
    image_path = "first_page.jpg"
    image.save(image_path)

    output = model.inference(image=image_path, prompt=task_prompt)["predictions"][0] 
    return output

task_name = "SGSInvoice"
task_prompt = f"<s_{task_name}>"
model = DonutModel.from_pretrained("uartimcs/donut-invoice-extract")
model.eval()


interface = gr.Interface(
    fn=analysis_process,
    inputs=gr.File(label="Upload PDF"),
    outputs="json"
)

interface.launch()