metadata
library_name: transformers
license: mit
language:
- ru
base_model:
- Akajackson/donut_rus
- naver-clova-ix/donut-base
pipeline_tag: image-to-text
The Donut (end-to-end transformer) model for text recognition
Using:
from transformers import AutoTokenizer, AutoModelForImageTextToText, AutoProcessor
from datasets import load_dataset
processor = AutoProcessor.from_pretrained("intexcp/donut")
tokenizer = AutoTokenizer.from_pretrained("intexcp/donut")
model = AutoModelForImageTextToText.from_pretrained("intexcp/donut")