Azure AI Document Intelligence
An Azure service that turns documents into usable data. Previously known as Azure Form Recognizer.
1,829 questions
Document Intelligence Studio - Custom Extraction Model, Training failed
![](data:image/svg+xml, %3Csvg xmlns='http://www.w3.org/2000/svg' height='64' class='font-weight-bold' style='font: 600 30.11764705882353px "SegoeUI", Arial' width='64'%3E%3Ccircle fill='hsl(44.800000000000004, 9%, 14%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3EPV%3C/text%3E%3C/svg%3E)
Pavith Vickneswararajah
0
Reputation points
Hello,
I'm trying to create and train a custom extraction model using my own labeled training dataset. My training data consists of approximately 5,000 PDF files, 5,000 labels.json files, and 5,000 ocr.json files. I'm currently on the S0 Tier (Pay-As-You-Go), but whenever I start a training process, it either runs indefinitely with the status "Running," or it fails with the error "InternalServerError: An unexpected error occurred."
I have tried starting the training process both through the Document Intelligence UI and via API calls, but neither method works. Do you have any tips on how to proceed?
{count} votes
-
![](data:image/svg+xml, %3Csvg xmlns='http://www.w3.org/2000/svg' height='64' class='font-weight-bold' style='font: 600 30.11764705882353px "SegoeUI", Arial' width='64'%3E%3Ccircle fill='hsl(281.6, 88%, 37%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3ESA%3C/text%3E%3C/svg%3E)
Saideep Anchuri 705 Reputation points Microsoft Vendor2024-12-30T11:01:33.2666667+00:00 Welcome to Microsoft Q&A Forum, thank you for posting your query here!
Based on the error message you received; it seems that an unexpected error occurred while training your custom model. This error message is not very specific, so it can be difficult to determine the exact cause of the issue. However, there are a few potential solutions that you can try to resolve this error:
Your training size is bit higher then allowed limit. Max Pages that can be included for Custom extraction model training is 500 pages with max size of 50 MB for all files. Could you check with 500 pages as total training file size.
If you still face the issue, I request you to raise a support case through Azure portal.
Kindly refer Below documentation: https://learn.microsoft.com/en-us/azure/ai-services/document-intelligence/train/custom-template?view=doc-intel-4.0.0#input-requirements
Thank You.
-
Saideep Anchuri 705 Reputation points Microsoft Vendor
2024-12-31T04:12:21.7633333+00:00 -
Saideep Anchuri 705 Reputation points Microsoft Vendor
2025-01-01T13:45:40.7933333+00:00 We haven’t heard from you on the last response and was just checking back to see if the give response was helpful.
Thank You.
-
Pavith Vickneswararajah 0 Reputation points
2025-01-02T10:40:18.4666667+00:00 hi @Saideep Anchuri , the response was not helpful but Im in contact with the support now.
Sign in to comment
Sign in to answer