qurator-spk/ocrd-galley
1
0
Fork 0
mirror of https://github.com/qurator-spk/ocrd-galley.git synced 2025-10-30 18:54:14 +01:00
Code Issues Projects Releases Wiki Activity

Refactor: Extract functions for the steps

Browse source
This commit is contained in:
Gerber, Mike 2019-06-19 12:51:52 +02:00
parent a2d547b857
commit c207859bcd
1 changed files with 27 additions and 19 deletions
Download patch file Download diff file Expand all files Collapse all files

46
my_ocrd_workflow
View file

@ -13,27 +13,35 @@ remove_filegrp() {
-d "//mets:fileGrp[@USE='$filegrp_use']" $mets -d "//mets:fileGrp[@USE='$filegrp_use']" $mets
} }
do_fontident() {
ocrd_typegroups_classifier_parameters='
{
"network": "/home/mike/devel/OCR-D/monorepo/ocrd_typegroups_classifier/ocrd_typegroups_classifier/models/classifier.tgc",
"stride":143
}'
remove_filegrp OCR-D-FONTIDENT mets.xml
ocrd-typegroups-classifier -l DEBUG \
-m mets.xml -I OCR-D-IMG -O OCR-D-FONTIDENT \
-p <(echo $ocrd_typegroups_classifier_parameters)
# XXX does DEFAULT have any meaning? /buerger_gedichte_1778.ocrd does not have
# any DEFAULT, yet -I DEFAULT seems to work for ocrd-typegroups-classifier
#ocrd workspace validate mets.xml
# XXX <error>Unspecified USE category 'FONTIDENT' in fileGrp 'OCR-D-FONTIDENT'</error>
# XXX <error>File 'OCR-D-FONTIDENT_OCR-D-IMG_0002' does not manifest any physical page.</error>
# XXX <notice>Won't download remote image <OCR-D-IMG/OCR-D-IMG_0002></notice>
}
ocrd_typegroups_classifier_parameters=' do_linesegmentation() {
{ remove_filegrp OCR-D-SEG-LINE mets.xml
"network": "/home/mike/devel/OCR-D/monorepo/ocrd_typegroups_classifier/ocrd_typegroups_classifier/models/classifier.tgc", ocrd-tesserocr-segment-line -l DEBUG \
"stride":143 -m mets.xml -I OCR-D-IMG -O OCR-D-SEG-LINE
}' #ocrd workspace validate mets.xml
remove_filegrp OCR-D-FONTIDENT mets.xml }
ocrd-typegroups-classifier -l DEBUG \
-m mets.xml -I OCR-D-IMG -O OCR-D-FONTIDENT \
-p <(echo $ocrd_typegroups_classifier_parameters) do_fontident
# XXX does DEFAULT have any meaning? /buerger_gedichte_1778.ocrd does not have do_linesegmentation
# any DEFAULT, yet -I DEFAULT seems to work for ocrd-typegroups-classifier
#ocrd workspace validate mets.xml
# XXX <error>Unspecified USE category 'FONTIDENT' in fileGrp 'OCR-D-FONTIDENT'</error>
# XXX <error>File 'OCR-D-FONTIDENT_OCR-D-IMG_0002' does not manifest any physical page.</error>
# XXX <notice>Won't download remote image <OCR-D-IMG/OCR-D-IMG_0002></notice>
remove_filegrp OCR-D-SEG-LINE mets.xml
ocrd-tesserocr-segment-line -l DEBUG \
-m mets.xml -I OCR-D-IMG -O OCR-D-SEG-LINE
#ocrd workspace validate mets.xml
# vim:tw=120: # vim:tw=120: