Use captioning

handler.py

@@ -5,12 +5,15 @@ from PIL import Image
 import pdb
 import requests
 
+MODEL = "google/pix2struct-screen2words-large" 
+
 class EndpointHandler():
     def __init__(self, path=""):
         #self.processor = AutoProcessor.from_pretrained("jasper-lu/pix2struct_embedding")
         #self.model = MarkupLMModel.from_pretrained("jasper-lu/pix2struct_embedding")
-        self.processor = AutoProcessor.from_pretrained("google/pix2struct-textcaps-base")
-        self.model = Pix2StructVisionModel.from_pretrained("google/pix2struct-textcaps-base")
+        self.processor = AutoProcessor.from_pretrained(MODEL)
+        self.processor.image_processor.is_vqa = False
+        self.model = Pix2StructVisionModel.from_pretrained(MODEL)
 
     def __call__(self, data: Any) -> List[List[Dict[str, float]]]:
         url = data.pop("inputs", data)