ankur-bohra commited on
Commit
e853e36
1 Parent(s): c52d6b0

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +12 -1
app.py CHANGED
@@ -43,9 +43,20 @@ def process_document(image, question):
43
  description = "Gradio Demo for Donut, an instance of `VisionEncoderDecoderModel` fine-tuned on DocVQA (document visual question answering). To use it, simply upload your image and type a question and click 'submit', or click one of the examples to load them. Read more at the links below."
44
  article = "<p style='text-align: center'><a href='https://arxiv.org/abs/2111.15664' target='_blank'>Donut: OCR-free Document Understanding Transformer</a> | <a href='https://github.com/clovaai/donut' target='_blank'>Github Repo</a></p>"
45
 
 
 
 
 
 
 
 
 
 
 
 
46
  demo = gr.Interface(
47
  fn=process_document,
48
- inputs=["image", "text"],
49
  outputs="json",
50
  title="Demo: Donut 🍩 for DocVQA",
51
  description=description,
 
43
  description = "Gradio Demo for Donut, an instance of `VisionEncoderDecoderModel` fine-tuned on DocVQA (document visual question answering). To use it, simply upload your image and type a question and click 'submit', or click one of the examples to load them. Read more at the links below."
44
  article = "<p style='text-align: center'><a href='https://arxiv.org/abs/2111.15664' target='_blank'>Donut: OCR-free Document Understanding Transformer</a> | <a href='https://github.com/clovaai/donut' target='_blank'>Github Repo</a></p>"
45
 
46
+ vqa_questions = {}
47
+ vqa_questions["ACCOUNT/BILL NUMBER"] = "What is the account or bill number?"
48
+ vqa_questions["TOTAL"] = "What is the total amount or total price?"
49
+ vqa_questions["ITEMS"] = "What are the items?"
50
+ vqa_questions["GST AMOUNT"] = "What is the GST or tax amount?"
51
+ vqa_questions["GST NO."] = "What is the GST number?"
52
+ vqa_questions[
53
+ "SELLER/BILLING DETAILS"
54
+ ] = "What are the seller details or billing details"
55
+ vqa_questions["BILLING ADDRESS"] = "What is the billing address?"
56
+
57
  demo = gr.Interface(
58
  fn=process_document,
59
+ inputs=["image"] + [gr.components.Textbox(value=question) for question in vqa_questions.values()],
60
  outputs="json",
61
  title="Demo: Donut 🍩 for DocVQA",
62
  description=description,