Commit graph

1245 commits

Author SHA1 Message Date
kba
dbe06867a6 wip: remove textline_light=True from call to EynollahXmlWriter 2025-12-10 14:24:32 +01:00
vahidrezanezhad
58000069cf Restore correct execution of export_textline_images_and_text 2025-12-10 13:14:38 +01:00
vahidrezanezhad
5716262629 Fix eynollah ocr --help so it works again 2025-12-10 13:14:38 +01:00
vahidrezanezhad
86d437b77b Restored correct functionality of the extract_only_images mode and cleaned up the argument handling 2025-12-10 13:14:38 +01:00
kba
4175b52768 log to STDERR not STDOUT 2025-12-10 13:14:38 +01:00
kba
04d21b9d92 🔥 refactor eynollah ocr
.
2025-12-10 13:14:38 +01:00
kba
244847e3d4 move line-gt extraction out of ocr to eynollah-training 2025-12-10 13:14:38 +01:00
kba
058478baf3 CI: do not upgrade (now-unpineed) torch 2025-12-10 13:14:38 +01:00
kba
fcd87fc3cf 💀 remove dead code from eynollah.py 2025-12-10 13:14:32 +01:00
kba
1eef5514d7 eynollah.py: fix kwargs to writer 2025-12-10 12:57:56 +01:00
kba
b7d3a6724b enforce kwargs for writer.build_... 2025-12-10 12:57:53 +01:00
kba
97959869ba remove more branches after textline_light default true 2025-12-10 12:57:03 +01:00
kba
5d497b0f72 factor out extract_only_images as eynollah extract-images 2025-12-10 12:57:01 +01:00
kba
b10773aae6 🔥 replace light_version/textline_light with True 2025-12-10 12:56:01 +01:00
kba
ca83cf934d fix imports from src/cli/cli_*/*_cli 2025-11-26 20:48:14 +01:00
kba
095b36c389 models: split into layout, extra and ocr
layout: Everything not OCR or extra
ocr: trocr/cnnrnn models
extra: obsolete or niche models
2025-11-26 19:49:59 +01:00
kba
000af16a47 🔥 remove torch pinning 2025-11-26 19:23:49 +01:00
kba
e503c1a0b7 drop obsolete multi-model binarization 2025-11-26 18:51:41 +01:00
kba
82266f8234 reorganize cli 2025-11-26 18:51:20 +01:00
kba
5a1900e664 🔥 remove OCR option from eynollah layout 2025-11-26 18:12:03 +01:00
kba
0f410c2e7c disable tf/keras logging on first import 2025-11-26 16:37:54 +01:00
kba
9d9d32daed update OCR-D bindings 2025-11-26 16:20:27 +01:00
kba
103c007368 . 2025-11-26 14:37:00 +01:00
kba
0149147e95 . 2025-11-25 13:45:47 +01:00
kba
67003b837c . 2025-11-13 16:56:04 +01:00
kba
d66549012f . 2025-11-13 14:57:28 +01:00
kba
b9bc8e79c0 github ci: cache models with model_zoo default config as key 2025-11-13 13:58:38 +01:00
kba
b34329dd61 tests: more path fixes 2025-11-13 12:21:48 +01:00
kba
9aeff6d155 tests: typo 2025-11-13 11:49:09 +01:00
kba
a72be69958 tests: fix model download URL 2025-11-13 11:48:23 +01:00
kba
3afbce023d tests: adapt paths 2025-11-13 11:46:31 +01:00
vahidrezanezhad
ed5b5c13dd Add test images; call TrOCR processor from the same directory as the TrOCR model 2025-11-07 12:47:21 +01:00
kba
8732007aaf . 2025-11-06 16:33:39 +01:00
kba
f902756ce1 try importing torch, then shapely, then tensorflow 2025-11-06 13:10:35 +01:00
kba
44037bc05d add layout marginalia test 2025-11-06 12:42:57 +01:00
kba
d224b0f7e8 try with shapely.set_precision(...mode="keep_collpased") 2025-11-06 11:55:40 +01:00
kba
0d84e7da16 Merge remote-tracking branch 'origin/docs_and_minor_fixes' into model-zoo
# Conflicts:
#	README.md
#	train/README.md
2025-11-06 11:37:10 +01:00
kba
53e879e289 make *test: another typo; 2025-11-05 16:19:55 +01:00
kba
e449dbab6d make *test: fix paths 2025-11-05 15:28:41 +01:00
kba
0bef6e297b make models: unzip to the versioned directory 2025-11-05 15:19:16 +01:00
kba
2c211095d7 make deps-test should not depend on the models 2025-11-05 15:02:55 +01:00
kba
b6c7283b4d further debugging 2025-11-05 14:41:18 +01:00
cneud
f90259d6e2 fix docs links 2025-10-30 22:24:54 +01:00
cneud
d5b7089bad Merge branch 'docs_and_minor_fixes' of https://github.com/qurator-spk/eynollah into docs_and_minor_fixes 2025-10-30 22:17:41 +01:00
cneud
9dbac280cc Revert "remove unnecessary backslash"
This reverts commit f212ffa22d.
2025-10-30 22:16:53 +01:00
cneud
2d35a0598d Revert "replace list declaration with list literal (faster)"
This reverts commit 9733d575bf.
2025-10-30 22:16:48 +01:00
cneud
70d8577a15 Revert "remove redundant parentheses"
This reverts commit 20a95365c2.
2025-10-30 22:16:41 +01:00
Clemens Neudecker
c9efbe1871
refactor image layout in examples.md 2025-10-30 16:52:59 +01:00
kba
8782ef17b2 CI: 🔥 upgrade torch for debugging 2025-10-30 12:19:35 +01:00
kba
62d05917c5 test_layout: str(Path) 2025-10-30 12:17:38 +01:00