search

DDMAL / text_alignment

Aligns correct transcripts to text images using a "messy" OCR and Needleman-Wunsch sequence alignment
MIT License
6 stars 1 forks source link

Switch to using Python 2's tempfile builtin library for managing Ocropus data #6

Closed timothydereuse closed 5 years ago

timothydereuse commented 5 years ago

Currently the OCR is being done in manually created folders which will persist and clutter up the docker container if something crashes before they can be cleaned up.

timothydereuse commented 5 years ago

Looks like tempfile / temp directories might not work in rodan. For now, the entirety of the OCR portion is within a try-except block that will always delete the manually created directory whether or not something goes wrong.