ADD : How to use

README.md

@@ -14,6 +14,28 @@ deplot_kr은 google의 pix2struct 구조를 기반으로 한 한국어 image-to-
 
 ## How to use
 
+You can run a prediction by input an image.
+Model predict the data table of text form in the image.
+
+```python
+from transformers import Pix2StructForConditionalGeneration, Pix2StructImageProcessor, AutoTokenizer, Pix2StructProcessor
+from PIL import Image
+
+image_processor = Pix2StructImageProcessor()
+tokenizer = AutoTokenizer.from_pretrained("brainventures/deplot_kr")
+processor = Pix2StructProcessor(image_processor=image_processor, tokenizer=tokenizer)
+
+model = Pix2StructForConditionalGeneration.from_pretrained("brainventures/deplot_kr")
+
+image_path = "IMAGE_PATH"
+image = Image.open(image_path)
+
+inputs = processor(images=image, return_tensors="pt")
+pred = model.generate(flattened_patches=flattened_patches, attention_mask=attention_mask, max_length=1024)
+print(processor.batch_decode(deplot_generated_ids, skip_special_token=True)[0])
+
+```
+
 
 
 ### Training data