You cannot select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
|
|
|
|
#!/bin/bash -x
|
|
|
|
|
set -e
|
|
|
|
|
|
|
|
|
|
self=`realpath $0`
|
|
|
|
|
self_dir=`dirname "$self"`
|
|
|
|
|
|
|
|
|
|
DATA_SUBDIR=data
|
|
|
|
|
get_from_web() {
|
|
|
|
|
download_to 'https://qurator-data.de/calamari-models/GT4HistOCR/model.tar.xz' 'calamari-models/GT4HistOCR/2019-07-22T15_49+0200'
|
|
|
|
|
download_to 'https://qurator-data.de/calamari-models/GT4HistOCR/2019-12-11T11_10+0100/model.tar.xz' 'calamari-models/GT4HistOCR/2019-12-11T11_10+0100'
|
|
|
|
|
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/2021-03-09/models.tar.gz' 'sbb_binarization/2021-03-09'
|
|
|
|
|
download_to --no-unpack 'https://qurator-data.de/mirror/github.com/tesseract-ocr/tessdata_best/archive/4.0.0-repacked.tar.gz' 'mirror/github.com/tesseract-ocr/tessdata_best/archive/4.0.0-repacked.tar.gz'
|
|
|
|
|
download_to --strip-components 1 'https://qurator-data.de/eynollah/models_eynollah.tar.gz' 'eynollah'
|
|
|
|
|
}
|
|
|
|
|
. $self_dir/qurator_data_lib.sh
|
|
|
|
|
handle_data
|
