Commits · 9b537f02d311507125aa743c413058003776d18d · Automatic Text Recognition / Data Generator

Feb 16, 2022
- Add style filter (handwritten, typewritten); support ignored_classes · 9b537f02
  Martin Maarand authored 3 years ago
  
  9b537f02
Feb 01, 2022
- no more best_classes · 3fbdf930
  Martin authored 3 years ago
  
  3fbdf930
- choose only one transcription based on the accepted worker version ids (order is important) · a5a40182
  Martin authored 3 years ago
  
  a5a40182
Nov 24, 2021
- add username to default cache path to avoid conflicts with other user caches · 5d328548
  Martin authored 3 years ago
  
  5d328548
Nov 05, 2021
- fix linting · 32de35af
  Martin authored 3 years ago
  
  32de35af
- add cached api client · 7c56a81c
  Martin authored 3 years ago
  
  7c56a81c
Oct 29, 2021
- Raise an exception if multiple transcriptions from the same text_line · d6cf517d
  Martin Maarand authored 3 years ago
  
  d6cf517d
- add skew extraction modes · 9ab6085f
  Martin Maarand authored 3 years ago
  
  9ab6085f
Oct 08, 2021
- cache full size images to make using several extractions of the same images faster · a297c651
  Martin Maarand authored 3 years ago
  
  a297c651
Aug 06, 2021
- remove deprecated filtering by source.slug · 641d6c3f
  Martin authored 3 years ago
  
  641d6c3f
Aug 05, 2021
- Don't filter vertical lines with rotation class · 681f7ff5
  Martin Maarand authored 3 years ago
  
  681f7ff5
Aug 03, 2021
- Polygon resize · 9f61d8c8
  Martin Maarand authored 3 years ago
  
  9f61d8c8
Jul 01, 2021
- fix color arg (before always grayscale) · f096d85d
  Martin Maarand authored 3 years ago
  
  f096d85d
Jun 18, 2021
- Use rotation class · 47be8292
  Martin Maarand authored 3 years ago
  
  47be8292
Jun 17, 2021
- forgot to commit main · 3353a171
  Martin authored 3 years ago
  
  3353a171
- rename script to avoid clash with package name · a53f8f96
  Martin Maarand authored 3 years ago
  
  a53f8f96
Jun 02, 2021
- Add deskew extraction · 540884e9
  Martin Maarand authored 3 years ago
  
  540884e9
Jan 29, 2021
- add sorts to make the splits reproducible · fcfdfcbe
  Martin authored 4 years ago
  
  fcfdfcbe
Jan 08, 2021
- use text_line by default · ada7640d
  Martin authored 4 years ago
  
  ada7640d
- update gitignore · 389d8bb5
  Martin authored 4 years ago
  
  389d8bb5
- add codespell ignore lines file · 3e6e387a
  Martin authored 4 years ago
  
  3e6e387a
- add gitlab ci, first test · dceab604
  Martin authored 4 years ago
  
  dceab604
- move kaldi_data_generator to dir · d1c3fc0b
  Martin authored 4 years ago
  
  d1c3fc0b
- remove duplicate f.close() · 447a1551
  Martin authored 4 years ago
  
  447a1551
- fix kraken format with polygon · 63e24774
  kermorvant authored 4 years ago
  
  63e24774
- Merge branch 'filter_by_worker_version' into 'master' · 85af0496
  Martin Maarand authored 4 years ago
  
  filtering by worker version id See merge request teklia/kaldi_data_generator!6
  85af0496
- use manual instead of None for manual transcriptions · 493b10c8
  Martin authored 4 years ago
  
  493b10c8
- applied filter by worker version commit · a0c0910c
  Chaza Abdelwahab authored 4 years ago
  
  a0c0910c
- Merge branch 'skip_vertical_lines' into 'master' · 1ccedc3b
  Martin Maarand authored 4 years ago
  
  Add option to skip vertical lines See merge request teklia/kaldi_data_generator!4
  1ccedc3b
- Add option to skip vertical lines · 45c9e42e
  Chaza Abdelwahab authored 4 years ago
  
  45c9e42e
Dec 16, 2020
- Merge branch 'kraken_integration' into 'master' · 90175f84
  Martin Maarand authored 4 years ago
  
  integrating kraken See merge request teklia/kaldi_data_generator!2
  90175f84
- integrating kraken · b263ee7c
  kermorvant authored 4 years ago
  
  b263ee7c
Oct 23, 2020
- --dataset_name is not required for --split_only · c6363c89
  Martin authored 4 years ago
  
  c6363c89
Sep 17, 2020
- use latest version of arkindex-client · 5a280bd7
  Martin authored 4 years ago
  
  5a280bd7
Sep 15, 2020
- fix formatting · a24a0f08
  Martin authored 4 years ago
  
  a24a0f08
- add option to use existing split; add option to create a split from already downloaded lines · d8a643d0
  Martin authored 4 years ago
  
  d8a643d0
Sep 14, 2020
- support new transcriptions · 338fda47
  Martin authored 4 years ago
  
  338fda47
Apr 07, 2020
- Merge branch 'filter_elements' into 'master' · 8a0edebd
  Martin Maarand authored 4 years ago
  
  Filter elements according to their classes See merge request teklia-projects/kaldi_data_generator!1
  8a0edebd
- Filter elements according to their classes · 96c2464a
  Raphael Toumi authored 4 years ago
  
  96c2464a
Mar 31, 2020
- fix formatting · 9556bdbc
  Martin authored 4 years ago
  
  9556bdbc