Spaces:
Running
Running
File size: 792 Bytes
295c649 241cd82 295c649 241cd82 7391863 241cd82 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 |
from transformers import ViTFeatureExtractor, BertTokenizer, VisionEncoderDecoderModel, AutoTokenizer, AutoFeatureExtractor
import gradio as gr
model=VisionEncoderDecoderModel.from_pretrained("priyank-m/beta-OCR")
tokenizer = AutoTokenizer.from_pretrained("xlm-roberta-base")
feature_extractor = AutoFeatureExtractor.from_pretrained("facebook/vit-mae-large")
def run_ocr(image):
pixel_values = feature_extractor(image, return_tensors="pt").pixel_values
# autoregressively generate caption (uses greedy decoding by default )
generated_ids = model.generate(pixel_values, max_new_tokens=50)
generated_text = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)[0]
return generated_text
demo = gr.Interface(fn=run_ocr, inputs="image", outputs="text")
demo.launch() |