Internal Server Error In Form Recognizer Studio

Jitesh Mistry 21 Reputation points
2022-03-23T10:21:03.24+00:00

We had already trained the model and created a composed model, but currently, we are facing the issue, it gives the 500. it was working fine but facing issues from the past few days it gives internal server errors. Internal Server Error

Azure AI Document Intelligence
Azure AI Document Intelligence
An Azure service that turns documents into usable data. Previously known as Azure Form Recognizer.
1,365 questions
{count} votes

4 answers

Sort by: Most helpful
  1. Hussain Tinwala 16 Reputation points
    2022-05-07T11:25:02.937+00:00

    Hi all,

    I'm running into the same problem. This just stopped working on a composed model I've trained randomly. No changes were made to the training data or API calls/setup. Testing with a document via the API and the GUI result in 500 Internal Server Error.

    My documents are all under 1 MB in both the training data set and the test forms.

    Following this and hoping for a quick resolution on this.

    3 people found this answer helpful.
    0 comments No comments

  2. taha.sonmez 11 Reputation points
    2022-05-18T07:43:29.543+00:00

    @Ramr-msft any update on issue? I'm still experiencing..

    2 people found this answer helpful.
    0 comments No comments

  3. Emily Harper 111 Reputation points
    2022-03-25T19:32:27.05+00:00

    Hi guys,

    I don't know if this is useful, but the other day I had this same error when training and it was due to one of the documents in my batch being too large. I had one that was 15MB compared to the other 1MB or less. I had it fail like this 6 times but once I removed it from my storage account, my training went through.

    This might not be the same in your cases, but at least this was the cause of my issue and so wanted to share.

    Emily

    1 person found this answer helpful.
    0 comments No comments

  4. taha.sonmez 11 Reputation points
    2022-04-28T14:12:56.077+00:00

    Hi @Ramr-msft ,

    I have similar issue on both testing the model and using the API. This issue raised today. v3 is not really stable.

    0 comments No comments