I am running a Demo workflow on Intelligent OCR, including the Classify Document Scope Activity. On some of the documents processed (i.e. not all of the same type, some get classified OK), I am getting the “Index was outside the bounds of the array” in Data Extraction Scope Activity.
I reckon the exception is related to the preceding Classify Document activity, if the Classification result returns no value. When I change input in Data Extraction Scope from ClassificationResult (in argument) to DocumentTypeID (thus removing the classification from the loop), the Extraction Activity works just fine.
I am passing ClasResult(0) in the in argument.
Any ideas how to make it more reliable (by changing/adding/removing the Keywords in Classifier maybe?) or any other fix to this? My experiments gain no improvement.
I would prefer to avoid manual classification via if or try/catch if achievable.