AnhxtanhLab/vietnamese-trocr-large-handwritten-v2

model = VisionEncoderDecoderModel.from_pretrained('AnhxtanhLab/vietnamese-trocr-large-handwritten')
processor = TrOCRProcessor.from_pretrained("microsoft/trocr-large-handwritten")

# Move model to GPU if available
device = "cuda" if torch.cuda.is_available() else "cpu"
model.to(device)

def predict_text(image):
    # Process image and move it to the correct device
    pixel_values = processor(image, return_tensors="pt").pixel_values
    pixel_values = pixel_values.to(device)

    # Generate prediction
    generated_ids = model.generate(pixel_values)
    generated_text = processor.batch_decode(generated_ids, skip_special_tokens=True)[0]
    return generated_text