[QUERY] Form Recognizer custom models and JSON

Query/Question I am using Azure Form Recognizer custom model to extract line and table data from PDF files. For few scenarios it’s not able to extract all the rows of a table (i.e. If a table contains 20 rows, it’s extracting 19 rows and not able to extract last one). Is there any way so that I can compare extracted json values and file content to check if Form Recognizer is not able to extract all the rows.

Environment:

• Name and version of the Library package used: Azure.AI.FormRecognizer 3.1.1
• Hosting platform or OS and .NET runtime version (dotnet --info output for .NET Core projects): Windows 10 and .NET 5
• IDE and version : Microsoft Visual Studio Community 2019

[biswajitsamal59]

Great. So a good way to test that analyzing the custom model is returning what you want is to use the tool to Analyze a new document using your custom model. There is a button where you can download the JSON and see if all the information you need is there.

Let us know if that helps

I have checked that. The PDF file contains a table with 1 header row and 20 data rows. But the downloaded JSON file contains table with 1 header and 19 data rows. Even the Form Recogniser Layout service giving same result (Without using custom model only analysing using Layout service).

[msftbot[bot]]

Hi, we’re sending this friendly reminder because we haven’t heard back from you in 7 days. We need more information about this issue to help address it. Please be sure to give us your input. If we don’t hear back from you within 14 days of this comment the issue will be automatically closed. Thank you!