search

ocr-d-modul-2-segmentierung / page-segmentation

Pixel classifier for historic prints. Mirror of https://gitlab2.informatik.uni-wuerzburg.de/ls6/ocr4all-pixel-classifier. Report issues here.
Other
5 stars 2 forks source link

where is the frontend? #6

Open bertsky opened 2 years ago

bertsky commented 2 years ago

Repository names vs. package names are very confusing here. IIUC you have:

package repo
page-content-detection ocr4all-page-content
ocr4all-page-content-detection ?
ocr4all-segmentation ocr4all-segmentation
ocr4all-pixel-classifier page-segmentation
ocr4all-pixel-classifier-frontend ?

IIRC ocr4all-segmentation used to depend on ocr4all-pixel-classifier, but now depends on ocr4all-page-content-detection, which is different than the pkg name of ocr4all-page-content.

Also, the README here says:

For training and direct usage, install ocr4all-pixel-classifier-frontend. This package only contains the library code.

And then all the examples use that CLI. (No description of the API or the model training here.)

But where is that repo?

crater2150 commented 2 years ago

I'm not sure what ocr4all-page-content-detection and ocr4all-segmentation are about. If I remember correctly, that was an experimental tool that was never fully made ocrd-compatible.

The frontend was hosted on our university gitlab, I've now added a mirror in this github org. It contains examples of model training.

Note that no one is actively working on any of these repositories. I sometimes merge pull requests in my free time, but I don't have time for comprehensive API documentation.