Hi, form recognizer currently doesn't support the above scenario, sorry for any inconvenience.
--- *Kindly Accept Answer if the information helps. Thanks.*
This browser is no longer supported.
Upgrade to Microsoft Edge to take advantage of the latest features, security updates, and technical support.
Hi,
I'm sort of looking to do the opposite of Model Compose on Form Recognizer. While Model Compose allows you to combine multiple models and the classifier determines which model to call when you upload a document, I'm looking to call multiple models on the same form (I noticed accuracies we're much better on a "micro-model" structure where multiple models are responsible for parsing different tags in a page) and then hopefully have Azure stitch the results together.
Has anyone attempted this? Since I can't find anything online, my current path forward is to code this "assembler" using Python.
Hi, form recognizer currently doesn't support the above scenario, sorry for any inconvenience.
--- *Kindly Accept Answer if the information helps. Thanks.*