vahid
|
ffdc776192
|
hybrid cnn & transformer model is integrated
|
3 years ago |
Clemens Neudecker
|
0877199c68
|
importing types from click is enough
|
3 years ago |
vahidrezanezhad
|
792619e830
|
Update cli.py
|
3 years ago |
vahidrezanezhad
|
82e0783e01
|
fixing #30. Directory of model(s) is needed not model file name.
|
3 years ago |
Clemens Neudecker
|
aab51dbe82
|
Update ocrd-tool.json
|
3 years ago |
Clemens Neudecker
|
83c60d7b26
|
cleanup
|
3 years ago |
cneud
|
b7aeb1ae8b
|
use tensorflow Keras backend
|
3 years ago |
Robert Sachunsky
|
73b7b48dfb
|
ocrd-tool.json: add directory content-type for model
|
3 years ago |
Robert Sachunsky
|
416b5dd7e2
|
adapt to TF1 deprecation
|
4 years ago |
Robert Sachunsky
|
ad50ed287d
|
binarizer: reset TF session prior to predict()
|
4 years ago |
Robert Sachunsky
|
02423648ec
|
OCR-D CLI: encapsulate model loading into setup(), triggered by init
|
4 years ago |
Konstantin Baierer
|
8dd05064b2
|
📦 v0.0.8
|
4 years ago |
vahid
|
2c7cd84649
|
resolving error for inputs whcih have smaller scale than model patch
|
4 years ago |
Konstantin Baierer
|
f41cd58550
|
extract page_image also for operation_level region
|
4 years ago |
Konstantin Baierer
|
4a5544d985
|
📦 v0.0.7
|
4 years ago |
Konstantin Baierer
|
a3b1e721b2
|
ocrd-cli: fix indentation
|
4 years ago |
Konstantin Baierer
|
4ceed759ad
|
processor: self.resolve_resource model in addition to SBB_BINARIZE_DATA
|
4 years ago |
Konstantin Baierer
|
ea69494d2a
|
📦 v0.0.6
|
4 years ago |
Konstantin Baierer
|
c9e41727bf
|
📦 v0.0.5
|
4 years ago |
Lucas Sulzbach
|
45f1509dbc
|
sbb_binarize: Load each model only once
|
4 years ago |
Lucas Sulzbach
|
0f7d4c589a
|
sbb_binarize: Start new session only once
|
4 years ago |
Konstantin Baierer
|
b98d7f51ae
|
📦 v0.0.4
|
4 years ago |
vahidrezanezhad
|
f363063e75
|
Merge pull request #9 from OCR-D/envvar-continued
combine SBB_BINARIZE_DATA and model parameter
|
4 years ago |
Konstantin Baierer
|
6ebd6f0d45
|
📦 v0.0.3
|
4 years ago |
Konstantin Baierer
|
e46b89af85
|
Merge branch 'release-v0.0.2' into release-v0.0.3
|
4 years ago |
Konstantin Baierer
|
71b2801183
|
🐛 read from image_path, not image
|
4 years ago |
Konstantin Baierer
|
b10b49c6ce
|
📦 v0.0.2
|
4 years ago |
Clemens Neudecker
|
4d145cc905
|
Merge pull request #11 from OCR-D/altimage-comments
🐛 typo: comment{,s}, fix #8
|
4 years ago |
Konstantin Baierer
|
c4b63fca47
|
🐛 typo: comment{,s}, fix #8
|
4 years ago |
Konstantin Baierer
|
664a441a2d
|
glob *.h5 files only in model dir, fix #7
|
4 years ago |
Konstantin Baierer
|
b62b32c998
|
combine SBB_BINARIZE_DATA and model parameter
|
4 years ago |
Konstantin Baierer
|
0c31eaf6b1
|
ocrd cli: "model" parameter replaceable by SBB_BINARIZE_DATA envvar
|
4 years ago |
Konstantin Baierer
|
84d6b1f693
|
🐛 numpy.ndarray.as{_,}type
|
4 years ago |
Konstantin Baierer
|
6543e74deb
|
remove "patches" parameter, always use_patches
|
4 years ago |
Konstantin Baierer
|
2bc6ccc4c0
|
replace patches string comparison with use_patches boolean
|
4 years ago |
Konstantin Baierer
|
fad7b7aff1
|
rewrite "normal" CLI with click
|
4 years ago |
Konstantin Baierer
|
645ec4124a
|
ocrd-tool: disable "line" level-of-operation for now
|
4 years ago |
Konstantin Baierer
|
1fa581283c
|
SbbBinarizer: refactor (variable names, less instance-wide state)
|
4 years ago |
Konstantin Baierer
|
a1c8f6f465
|
line-level binarization independent of region-level
|
4 years ago |
Konstantin Baierer
|
e5bc5572a2
|
change description ht @bertsky @vahidrezanezhad
|
4 years ago |
Konstantin Baierer
|
b7a6301948
|
fix cv2pil ht @bertsky
|
4 years ago |
Konstantin Baierer
|
eece89b9ac
|
patches should be true by default, ht @vahidrezanezhad
|
4 years ago |
Konstantin Baierer
|
2eec6876fa
|
Append "binarized" to AlternativeImage/comments
Co-authored-by: Robert Sachunsky <38561704+bertsky@users.noreply.github.com>
|
4 years ago |
Konstantin Baierer
|
ee82d188bf
|
Set page_id from input file
Co-authored-by: Robert Sachunsky <38561704+bertsky@users.noreply.github.com>
|
4 years ago |
Konstantin Baierer
|
0650bad533
|
image_from_segment: exclude already binarized images
Co-authored-by: Robert Sachunsky <38561704+bertsky@users.noreply.github.com>
|
4 years ago |
Konstantin Baierer
|
12b44af329
|
convert between cv2 and pil, DRY binarizer call
|
4 years ago |
Konstantin Baierer
|
a4c0cf0a47
|
shut up keras/tensorflow
|
4 years ago |
Konstantin Baierer
|
d4c97ebf72
|
add OCR-D click interface
|
4 years ago |
Konstantin Baierer
|
ee26ebd7d8
|
implement region/line binarization
|
4 years ago |
Konstantin Baierer
|
5909f94fab
|
initial OCR-D interface
|
4 years ago |