InvalidContentSourceFormat: Invalid content source: Could not read build content.

Question

I've been able to train both a classifier and extractor model, but after restructuring the blob container so that the models could train out of separate subdirectories in the container, I found that the classifier training keeps failing with the error in the title. I can still label and see the files on the document intelligence app so the file path should be correct, so I'm not sure what is wrong.

Answer 1

Hello @Tyde Hashimoto , I'm glad that you were able to resolve your issue and thank you for posting your solution so that others experiencing the same thing can easily reference this! Since the Microsoft Q&A community has a policy that "The question author cannot accept their own answer. They can only accept answers by others ", I'll repost your solution in case you'd like to "Accept " the answer.

Issue: InvalidContentSourceFormat: Invalid content source: Could not read build content.

Solution: After verifying the filenames and found that one of the subfolders had an extra space which made it point to a subdirectory which didn't exist. After fixing that was able to train a new model.

Check the .jsonl files in Azure storage blob container, where it contains the name of each individual training files. Try to check the list if that contains any no-exist files and also the file naming details.

If you have any other questions or are still running into more issues, please let me know.
Thank you again for your time and patience throughout this issue.

Regards,
Vasavi

Please remember to "Accept Answer" if any answer/reply helped, so that others in the community facing similar issues can easily find the solution.