import gradio as gr
from transformers import TrOCRProcessor, VisionEncoderDecoderModel
import requests
from PIL import Image

processor = TrOCRProcessor.from_pretrained("paran3xus/typress_ocr")
model = VisionEncoderDecoderModel.from_pretrained('paran3xus/typress_ocr')

# load image examples
urls = ["https://huggingface.co/spaces/paran3xus/typress_ocr_space/resolve/main/test_img/1.png", "https://huggingface.co/spaces/paran3xus/typress_ocr_space/resolve/main/test_img/2.png", "https://huggingface.co/spaces/paran3xus/typress_ocr_space/resolve/main/test_img/3.png"]
for idx, url in enumerate(urls):
  image = Image.open(requests.get(url, stream=True).raw)
  image.save(f"image_{idx}.png")

def process_image(image):
    # prepare image
    pixel_values = processor(image, return_tensors="pt").pixel_values

    # generate (no beam search)
    generated_ids = model.generate(pixel_values)

    # decode
    generated_text = processor.batch_decode(generated_ids, skip_special_tokens=True)[0]

    return generated_text

title = "Interactive demo: Typress OCR"
description = "Demo for Typress OCR, an TrOCR model for Typst Mathematical Expressions Recognition. To use it, simply upload a image or use one of the example images below and click 'submit'. Results will show up in a few seconds."
article = "<p style='text-align: center'><a href='https://github.com/ParaN3xus/typress'>Github Repo</a></p>"
examples =[["image_0.png"], ["image_1.png"], ["image_2.png"]]

iface = gr.Interface(fn=process_image, 
                     inputs=gr.Image(type="pil"), 
                     outputs=gr.Textbox(),
                     title=title,
                     description=description,
                     article=article,
                     examples=examples)
iface.launch(debug=True)