search

ocropus-archive / DUP-ocropy

Python-based tools for document analysis and OCR
Apache License 2.0
3.41k stars 590 forks source link

complement of readme file #300

Closed xS-Sx closed 5 years ago

kba commented 5 years ago

Since this is GitHub, git clone is implied for all the install-from-source-steps IMHO.

But if #319 lands, we should consider adding steps to install with pip. In @OCR-D we've been publishing a fork to pypi, so you can do

pip install ocrd-fork-ocropy

without cloning and in fact use ocropy as a dependency in other projects.

zuphilip commented 5 years ago

Yes, the pip looks really simple. But will this also download the models? Is it clear where I have to save the new models if I have installed ocropy with pip locally or globally? (Maybe, we have to add more documentation then or can even provide some sort of helper script for such cases...)

kba commented 5 years ago

Users do have to save models to one of the places ocropy searches for them. They can do that manually (wget), use https://github.com/kba/ocr-models-client or we could bundle a script like kraken does.