LSTM Train data creation: list all fonts in --fontlist or run several individual commands
52 views
Skip to first unread message
Sim Tov
Aug 22, 2021, 11:03:18 AM8/22/21
to tesser...@googlegroups.com
Does it make any difference whether I run one command for all 8 fonts:
tesstrain.sh --fonts_dir Fonts --lang heb --linedata_only
--noextract_font_properties --langdata_dir ./langdata --tessdata_dir
/usr/share/tesseract-ocr/4.00/tessdata/ --output_dir output/train
--fontlist <font1> <font2> <...> <font8>
Or 8 separate commands:
tesstrain.sh --fonts_dir Fonts --lang heb --linedata_only
--noextract_font_properties --langdata_dir ./langdata --tessdata_dir
/usr/share/tesseract-ocr/4.00/tessdata/ --output_dir output/train
--fontlist <font1>
...
tesstrain.sh --fonts_dir Fonts --lang heb --linedata_only
--noextract_font_properties --langdata_dir ./langdata --tessdata_dir
/usr/share/tesseract-ocr/4.00/tessdata/ --output_dir output/train
--fontlist <font8>
?
Reply all
Reply to author
Forward
0 new messages