Skip to content

Model zoo #207

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Open
kba wants to merge 87 commits into
base: main
Choose a base branch
from
Open

Model zoo #207

kba wants to merge 87 commits into from

Conversation

@kba
Copy link
Contributor

@kba kba commented Oct 20, 2025 •
edited
Loading

Still WIP, intermediate state for discussion

  • all logic related to loading ML models moved to EynollahModelZoo
  • all models can have variants, such as light. Default is ''
  • model overrides are a three-string tuple (model_category, model_variant, model_filename)
  • Eynollah_ocr moved into a separate module

TODO:

  • better data structure for the defaults (including documentation e.g.)
  • apply to remaining runners (i.e. classes that have a run method and are called via CLI)
  • [ ] Factor out a common base class for all runners not required for this PR
  • make --model* options for the top-level eynollah click group

kba and others added 19 commits October 17, 2025 17:47
@kba
Refactor CLI for consistent logging and late imports
6c89888
@cneud
expand keywords and supported Python versions
2a1f892
@cneud
remove redundant parentheses
20a9536
@cneud
replace list declaration with list literal (faster)
9733d57
@cneud
remove unnecessary backslash
f212ffa
@cneud
readme and documentation updates
496a0e2
@vahidrezanezhad
updating model based reading orde detection
9d2dbb8
@vahidrezanezhad
Update flowchart
3ec5ceb
@vahidrezanezhad
updating heuristics and ocr documentation
c845537
@kba
factor model loading in Eynollah to EynollahModelZoo
a850ef3
@kba
adapt tests to -l being top-level option now
b90cfdf
@kba
move Eynollah_ocr to separate module
48d1198
@kba
organize imports mostly
d609a53
@kba
Introduce model_zoo to Eynollah_ocr
062f317
@cneud
combine Docker docs
6e3399f
@cneud
integrate training docs
e5254dc
@cneud
integrate ocrd docs
230e7cc
@cneud
small fixes to main readme
7d70835
@kba
cli: model -> model_basedir
44b75eb
kba
kba commented Nov 6, 2025
View reviewed changes
tests/cli_tests/test_layout.py
)
assert len(list(outdir.iterdir())) == 2

# def test_run_eynollah_layout_marginalia(
Copy link
Contributor Author

@kba kba Nov 6, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@bertsky This regression test is from a discussion with Vahid where he noticed that v0.5.0 produces 5 marginalia regions as expected but none in v0.6.0+.

Disabled for now until the segfaults have been fixed.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

No reviews

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

4 participants

@kba @cneud @vahidrezanezhad