vc1492a/tidd

Improve speed of dataset (image) generation code

vc1492a opened this issue · 1 comments

The script that generates the images for use in modeling is very slow and could be sped up. Not a high priority though because we could limit the data shared as part of a paper / on github to only the pre-prepared images.

This has been addressed in the feature/refactor branch through parallel processing. The speed is far faster than before when using multiple CPU cores for generating images used in model training and testing.