Marcel a99afef319 fix(training): only count reviewed blocks as checked text for recognition ...

Previously all MANUAL blocks counted as eligible training data, even ones
where text was filled in by guided OCR but never explicitly reviewed. This
caused segmentation and recognition counts to always match.

Now only reviewed=true blocks qualify for recognition training, so the
counts properly reflect: segments = all drawn annotation boxes,
checked text = only boxes where the user has verified the transcription.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

2026-04-13 18:00:59 +02:00

..

.mvn/wrapper

build: add mvn properties

2026-03-17 13:33:02 +00:00

api_tests

restructure: flatten workspace nesting, move devcontainer to root

2026-03-15 11:47:58 +01:00

src

fix(training): only count reviewed blocks as checked text for recognition

2026-04-13 18:00:59 +02:00

.gitignore

restructure: flatten workspace nesting, move devcontainer to root

2026-03-15 11:47:58 +01:00

Dockerfile

fix(backend): switch base image from alpine to debian for bash compatibility

2026-03-19 12:03:14 +01:00

HELP.md

restructure: flatten workspace nesting, move devcontainer to root

2026-03-15 11:47:58 +01:00

mvnw

restructure: flatten workspace nesting, move devcontainer to root

2026-03-15 11:47:58 +01:00

mvnw.cmd

build: add mvn properties

2026-03-17 13:33:02 +00:00

pom.xml

feat(training): add segmentation training pipeline and complete Part 6

2026-04-13 15:15:17 +02:00