Hello Het Patel,
Welcome to the Microsoft Q&A and thank you for posting your questions here.
I understand that you are using "Custom extraction model" in Azure Document Intelligence Studio, and you need to know how you could train custom classification model (template) for single invoices that spans across pages.
The error shows you're using same field name across 2 pages though, you blurred the names.
However, this is a known issue with cross-page labeling in Azure Document Intelligence Studio. My tips and advice to resolve this are the followings:
- Follow the latest guidelines from the Azure Document Intelligence documentation. - https://learn.microsoft.com/en-us/azure/ai-services/document-intelligence/concept-custom-label?view=doc-intel-4.0.0
- When dealing with tables that span multiple pages, you need to label each row of the table across the different pages in a single table. Your dataset should include samples where tables span multiple pages and label them consistently. This is a similar question answered by @dupammi on this platform.
- Using a custom neural model instead of a template model are better suited for handling complex scenarios like tables spanning multiple pages.
- You can also implement post-processing logic to merge table rows that are split across pages. This can help in cases where the model extracts partial rows from different pages.
- Check that the fields are uniquely named and that you are not inadvertently using the same field name across multiple pages.
I hope this is helpful! Do not hesitate to let me know if you have any other questions.
Please don't forget to close up the thread here by upvoting and accept it as an answer if it is helpful.