Spaces:
Runtime error
Runtime error
import gradio as gr | |
from transformers import TrOCRProcessor, VisionEncoderDecoderModel | |
import requests, re, base64 | |
from PIL import Image | |
from io import BytesIO | |
processor = TrOCRProcessor.from_pretrained("microsoft/trocr-small-printed") | |
model = VisionEncoderDecoderModel.from_pretrained("pENrknSoysneed/8kun-captcha-ocr-meme") | |
# # load image examples | |
# urls = [ | |
# 'https://storage.googleapis.com/trocr-captcha.appspot.com/captcha_images_v2/nfcb5.png', | |
# 'https://storage.googleapis.com/trocr-captcha.appspot.com/captcha_images_v2/p57fn.png', | |
# 'https://storage.googleapis.com/trocr-captcha.appspot.com/captcha_images_v2/w2yp7.png', | |
# 'https://storage.googleapis.com/trocr-captcha.appspot.com/captcha_images_v2/pme86.png', | |
# 'https://storage.googleapis.com/trocr-captcha.appspot.com/captcha_images_v2/w4nfx.png', | |
# 'https://storage.googleapis.com/trocr-captcha.appspot.com/captcha_images_v2/nf8b8.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 | |
image_data = re.sub('^data:image/.+;base64,', '', image) | |
im = Image.open(BytesIO(base64.b64decode(image_data))) | |
#Take's the picture | |
pixel_values = processor(im, 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 = "8kun captcha solver 1 in 8" | |
description = "Due to events. in 8chan staff moderation. I am attacking it. The gamergate shitposting days are over. and so is 8chan." | |
# article = "<p style='text-align: center'><a href='https://arxiv.org/abs/2109.10282'>TrOCR: Transformer-based Optical Character Recognition with Pre-trained Models</a> | <a href='https://github.com/microsoft/unilm/tree/master/trocr'>Github Repo</a></p>" | |
# examples =[["image_0.png"], ["image_1.png"], ["image_2.png"], ["image_3.png"], ["image_4.png"], ["image_5.png"]] | |
#css = """.output_image, .input_image {height: 600px !important}""" | |
iface = gr.Interface(fn=process_image, | |
# inputs=gr.inputs.Image(type="pil"), | |
inputs=gr.Textbox(placeholder="base64 string (right-click => copy-link) ..."), | |
outputs=gr.outputs.Textbox(), | |
title=title, | |
description=description, | |
# article=article, | |
# examples=examples | |
) | |
iface.launch(debug=True) |