I was wondering why the OCR never worked despite the jobs saying to have finished (and many reddit users saying the same)
I looked at the logs, and oh boy, I should have looked earlier Apart of whats in the text file... suddenly then it starts going through 28k images, all failing with:
see comments
see comments
I dont understand what's the problem My network has access to the www, for example ML models can download just fine... but this one fails Everything is up to date.