app.py

Test Image-Text to Text Models

app.py

@@ -0,0 +1,11 @@
+import streamlit as st
+from transformers import AutoProcessor, UdopForConditionalGeneration
+from datasets import load_dataset
+
+processor = AutoProcessor.from_pretrained("microsoft/udop-large", apply_ocr=True)
+model = UdopForConditionalGeneration.from_pretrained("microsoft/udop-large")
+
+# encoding = processor(image, question, words, boxes=boxes, return_tensors="pt")
+
+# predicted_ids = model.generate(**encoding)
+# print(processor.batch_decode(predicted_ids, skip_special_tokens=True)[0])