@Ioana_Gligan Since I was modeling your sample code that you posted previously, could you please take a look at this post and let me know if you have any thoughts? Really appreciate this as it is time sensitive.
I have successfully implemented the Document Understanding model including Create Document Validation and Wait for Validation activities. This works well for a single invoice, but this is not practical at all in real life commercial A/P departments.
I found a sample workflow that using parallel processing to manage multiple documents. When I tried it, I got the error message below. The first invoice made it through as an Action, but the 2nd occurrence caused an error.
I am trying to build a demo for a prospect for the middle of the week.
The automatically determined extracted results references do not match the Document Object Model. They may have been created using a different Document Object Model.
at UiPath.IntelligentOCR.Activities.BaseOrchestratorClientAsyncActivity.ThrowIfNeeded(Task task, Boolean suppressThrowException)
at UiPath.IntelligentOCR.Activities.BaseOrchestratorClientAsyncActivity.EndExecute(AsyncCodeActivityContext context, IAsyncResult result)
at System.Activities.AsyncCodeActivity.System.Activities.IAsyncCodeActivity.FinishExecution(AsyncCodeActivityContext context, IAsyncResult result)
at System.Activities.AsyncCodeActivity.CompleteAsyncCodeActivityData.CompleteAsyncCodeActivityWorkItem.Execute(ActivityExecutor executor, BookmarkManager bookmarkManager)
You are probably using variables within the parallel for each loop that are scoped outside the parallel for each and they get overwritten in two parallel branches. Try to set your variable scope within the body of the parallel for each (for the variables that are specific to each document being processed).
Hello Suraj,
Not multiple invoices on single pdf but different pdf’s in parallel for each so that presents data validation in Action Centre can be done simultaneously for all pdf’s
Hi @suraj.setty , I want to process multiple invoices present on a single pdf. each invoice is of same supplier and format, but can be one page or multiple pages.
You can train invoice with different variations of pages in a machine learning model and use a machine learning info extractor. This way you can extract all pages with the same taxonomy.