You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I see the need for at least two more elaborate all-in-one tasks we could encapsulate as well:
automatic tests (e.g. make tests): This would delegate to the modules' respective make test or whatever is available. We would sometimes have to install additional dependencies like make deps-test or clone more subrepos.
download models (e.g. make models): This would delegate to the modules' respective model files or other data. In some cases this means cloning more subrepos, in others running web downloads.
Both would have to be encapsulated locally in each module section. When used in Dockerfiles during build-time these would produce larger development images.
The text was updated successfully, but these errors were encountered:
What about components for training? Adding make training-install for Tesseract would be easy and provide tools not only for training, but also for handling of traineddata (for example adding word lists). A new tesstrain submodule (from tesseract-ocr/tesstrain) would be needed for real training.
I see the need for at least two more elaborate all-in-one tasks we could encapsulate as well:
make tests
): This would delegate to the modules' respectivemake test
or whatever is available. We would sometimes have to install additional dependencies likemake deps-test
or clone more subrepos.make models
): This would delegate to the modules' respective model files or other data. In some cases this means cloning more subrepos, in others running web downloads.Both would have to be encapsulated locally in each module section. When used in Dockerfiles during build-time these would produce larger development images.
The text was updated successfully, but these errors were encountered: