ML skill - field ID

Hi,

I created a data labeling session in document understanding, however the default field id is picking in the ml extractor and not the updated one. is there anyway of sorting this out or do I have to recreate the skill?

Hi @Aki1111

Are you referring to the item field here? If yes, the items field is just an indication that you are dealing with the tabular data & it shall be mapped directly with the items itself. Whatever comes under the items range (Custom columns as you have indicated) will be considered for extraction purpose, items have nothing to do with the actual extraction.

Hope this helps,
Best Regards.

Hi I understand but I need it to be mapped like how I labeled it in Document understanding, is there any way to update it in followings headings.

@Aki1111

Can you please specify how many documents you have labeled in order to extract custom field data like end-date, consumption, etc?

Best Regards.

You have to redeploy the skill after exporting the data with changed column name!

20 documents I have labeled

okay I will delete the skill and redeploy it and provide you with feedback if that worked.

@Aki1111

For a model to extract the data perfectly, you need to label a certain number of documents, as training is a resource-hungry action. When it comes to column data, it’s even more-

Column fields (item unit price, item quantity)

  • For Column fields, you need at least 50-200 document samples per column field, so for 5 column fields, with clean and simple layouts you might get good results with 300 document samples. For highly complex and diverse layouts, it might require over 1000 document samples. To cover multiple languages, then you need at least 200-300 document samples per language, assuming they cover all the different fields. So, for 10 header fields and 4 column fields with 2 languages, 600 document samples might be enough (400 for the columns and headers, plus 200 for the additional language), but in some cases might require 1200 or more document samples.

https://docs.uipath.com/document-understanding/automation-cloud/latest/user-guide/training-high-performing-models

In your case, you can try labeling more documents in order to make the model stable.

Hope this helps,
Best Regards.

it worked, Thank you.

Thank you for your help, I deployed the ml skill again.

This topic was automatically closed 3 days after the last reply. New replies are no longer allowed.