mirror of
https://github.com/qurator-spk/ocrd-galley.git
synced 2025-07-01 01:19:52 +02:00
🚧 Add preliminary support for ocrd_calamari's Calamari 1 version
This commit is contained in:
parent
f2ed0b5cbf
commit
b6bbc7ca3a
3 changed files with 34 additions and 2 deletions
|
@ -13,8 +13,8 @@ RUN ${PIP_INSTALL} \
|
|||
|
||||
|
||||
# Copy OCR models
|
||||
RUN mkdir -p /var/lib/calamari-models
|
||||
COPY data/calamari-models/GT4HistOCR /var/lib/calamari-models/GT4HistOCR
|
||||
RUN mkdir -p /var/lib/calamari-models/GT4HistOCR
|
||||
COPY data/calamari-models/GT4HistOCR/2019-07-22T15_49+0200 /var/lib/calamari-models/GT4HistOCR/2019-07-22T15_49+0200
|
||||
|
||||
|
||||
|
||||
|
|
26
Dockerfile-ocrd_calamari-feat-update-calamari1
Normal file
26
Dockerfile-ocrd_calamari-feat-update-calamari1
Normal file
|
@ -0,0 +1,26 @@
|
|||
FROM my_ocrd_workflow-core
|
||||
|
||||
|
||||
# XXX https://github.com/OCR-D/core/issues/642
|
||||
#ARG PIP_INSTALL="pip3 install --no-cache-dir --use-feature=2020-resolver"
|
||||
ARG PIP_INSTALL="pip3 install --no-cache-dir"
|
||||
|
||||
|
||||
# Build pip installable stuff
|
||||
RUN ${PIP_INSTALL} \
|
||||
https://github.com/OCR-D/ocrd_calamari/archive/feat/update-calamari1.zip
|
||||
|
||||
|
||||
# Copy OCR models
|
||||
RUN mkdir -p /var/lib/calamari-models/GT4HistOCR
|
||||
COPY data/calamari-models/GT4HistOCR/2019-12-11T11_10+0100 /var/lib/calamari-models/GT4HistOCR/2019-12-11T11_10+0100
|
||||
|
||||
|
||||
|
||||
# Check pip dependencies
|
||||
# XXX https://github.com/OCR-D/core/issues/642
|
||||
#RUN pip3 check
|
||||
|
||||
|
||||
# Default command
|
||||
CMD ["ocrd-calamari-recognize"]
|
6
build
6
build
|
@ -7,6 +7,7 @@ self_dir=`dirname "$self"`
|
|||
DATA_SUBDIR=data
|
||||
get_from_annex() {
|
||||
annex_get 'calamari-models/GT4HistOCR/2019-07-22T15_49+0200/*.ckpt*'
|
||||
annex_get 'calamari-models/GT4HistOCR/2019-12-11T11_10+0100/*.ckpt*'
|
||||
annex_get 'tesseract-models/GT4HistOCR/*.traineddata'
|
||||
annex_get 'textline_detection/*.h5'
|
||||
annex_get 'mirror/github.com/tesseract-ocr/tessdata_best/archive/4.0.0-repacked.tar.gz'
|
||||
|
@ -14,6 +15,7 @@ get_from_annex() {
|
|||
}
|
||||
get_from_web() {
|
||||
download_to 'https://qurator-data.de/calamari-models/GT4HistOCR/model.tar.xz' 'calamari-models/GT4HistOCR/2019-07-22T15_49+0200'
|
||||
# XXX Calamari 1 model
|
||||
download_to 'https://qurator-data.de/tesseract-models/GT4HistOCR/models.tar' 'tesseract-models/GT4HistOCR'
|
||||
download_to 'https://qurator-data.de/sbb_textline_detector/models.tar.gz' 'textline_detection'
|
||||
download_to --strip-components 1 'https://qurator-data.de/sbb_binarization/models.tar.gz' 'sbb_binarization'
|
||||
|
@ -32,3 +34,7 @@ docker build --cache-from=my_ocrd_workflow-sbb_textline_detector -t my_ocrd_work
|
|||
docker build --cache-from=my_ocrd_workflow-sbb_binarization -t my_ocrd_workflow-sbb_binarization -f Dockerfile-sbb_binarization .
|
||||
docker build --cache-from=my_ocrd_workflow-ocrd_cis -t my_ocrd_workflow-ocrd_cis -f Dockerfile-ocrd_cis .
|
||||
docker build --cache-from=my_ocrd_workflow-ocrd_fileformat -t my_ocrd_workflow-ocrd_fileformat -f Dockerfile-ocrd_fileformat .
|
||||
|
||||
|
||||
# XXX
|
||||
docker build --cache-from=my_ocrd_workflow-ocrd_calamari-feat-update-calamari1 -t my_ocrd_workflow-ocrd_calamari-feat-update-calamari1 -f Dockerfile-ocrd_calamari-feat-update-calamari1 .
|
||||
|
|
Loading…
Add table
Add a link
Reference in a new issue