Document Intelligence - ML Solution Error NSE0013 & NSE0016 - Vancouver

VT1308
Tera Contributor

Hello everyone!
I'm currently encountering an issue with Document Intelligence on a demo instance running in Vancouver. I've been working on setting up a use case, defining fields (keys), and creating a document task with attachments. The expectation is that the OCR and AI capabilities should kick in to extract the required fields from the document. 


However, I've run into a problem where my Document Task 
fails to process the task as shown in the attachment.

After investigating the logs and the ml_Solution.list, I found two errors shown as NSE0013 and NSE0016. Please refer to the attachments for more information.

I've tried searching for information regarding these errors, but I couldn't find anything specific. Any help or suggestions on how to tackle these errors would be much appreciated!

Best regards,
Vinh

7 REPLIES 7

Brian Bakker
ServiceNow Employee
ServiceNow Employee

@VT1308 

Do you have duplicate training records in table [ml_solution], as per KB1286265? Are there any other doc_intel solutions in the [ml_solution] table or is this the first time you are trying to use Document Intelligence?

@Brian Bakker thanks a lot for your response!
Unfortunately I have no access to your referred article (KB1286265) in order to verify your question. There are also other doc_intel solutions in the [ml_solution] table, but all of them have failed and show the same status.

Hope this information helps you! Please let me know in case you have questions or need more information.
Thank you and looking forward to your response!

Loic1
ServiceNow Employee
ServiceNow Employee

Hi Vinh,

You mentioned that you want to extract values from a document, but I see the ML Solution is related to Document Classifier, which is our new capability to classify documents, not extract.

Can you make sure that you create your use case from the Document Intelligence Admin page or from the menu Document Data Extraction?

If the Use Case is for Classification, then it is appropriate to use the Document Classifier > Use Case menu.

Hope this will help.

Loic

VT1308
Tera Contributor

Hi @Loic1,

thanks a lot for your help! I agree with you that my Use Case was created through the Document Classifier. The reason for this, is that I'm not able to create a new Use Case through the Document Intelligence Admin page. Once I try to create a new Use Case through the "New Use Case" Button, it refreshes the current page after submitting, but without creating the Use Case. The two use cases you can see on the screenshots are created through the Document Classifier.

Also the Document Intelligence page shows a weird collection of "Filters" as default setup. 
Please refer to the attachment for further information.

Did you ever face similiar issues or may have suggestions on how to tackle these issues?
Thanks a lot in advance!
Vinh