ML skill return diferent results in labaling and automation

Hello everyone,

I’m having an issue with Machine Learning Skill in a document.

When I try predicting it in the labeling, it extracts exactly “ISF (Importer Security Filling)”, as it was supposed to.

But when I try to extract it, from the same file in the automation, “(Importer” text is cutted off.

I checked the ML skill in the labeling and in the automation, and they are the same.

Any idea why this is happening?

thanks in advance.

Couple of things to check here.

Please check if your ML skill is the same as the one in your AI center. Check if both versioning is the same.

Next check if you have the same OCR endpoint in pre labeling & digitization. You need to have the same OCR to get the same results.

Please refresh your ML capabilities and test all these out again. Let me know if this works.