RuntimeError: Some images were not loaded.

[udit-pandey-1]

OS: Ubuntu - 20.04.6 LTS
Python: 3.11
Requirements:
unstructured==0.14.2 unstructured-inference==0.7.33 pillow-heif==0.16.0

I am getting the following error when extracting text and images from pdf:

The way I am using unstructured is:

Is there anyway we can fix this issue?

[tbs17]

Hi there, some of the troubleshooting steps we can consider:

Check PDF File Accessibility:

Ensure the PDF file is accessible and not corrupted.
Try opening the file with a PDF reader to verify its integrity.

File Path and Permissions:
Verify that the file path is correct.
Ensure you have the necessary permissions to read the file.

Temporary File Issues:

The error suggests a problem with accessing a temporary file. Ensure that the file is correctly created and accessible before processing it.
You might want to use a different temporary directory or explicitly manage the creation and deletion of temporary files.

Handling Large Files:

If the PDF is very large, it might cause issues during processing. Try with a smaller PDF to see if the problem persists.

Library Versions and Dependencies:

Ensure that all the libraries (unstructured, PyMuPDF, etc.) are up-to-date. There might be bug fixes or improvements in newer versions.

Debugging and Logging:

Add logging to your script to capture more details about where the error occurs.
Log the paths of the files being processed, and any other relevant information.

[udit-pandey-1]

Hi, thanks for the reply.
I got over the error by upgrading to unstructured-inference==0.7.33.
However, I am getting a different issue now and have opened an issue for it (Unstructured-IO/unstructured#3102).

I'll anyways close this one for now.