Azure Form Recognizer Extraction Issue. I have trained some PDFs with the help of Labelling Tool but it is not giving accurate results upon analysing?

Question

I have trained some PDFs with the help of Labelling Tool (FOTT) but it is not giving accurate results upon analyzing?

Answer 1

Hi, thanks for reaching out. Here's the official document for training custom model as well information on how to improve results. Here are some additional resources:

• Custom model input requirements
• Training data tips
• Best practices to improve custom model quality

Let us know if you have further questions after reviewing above resources. Thank!