Diachronic full normalization challenge
Do OCR (or OCR post-correction) and modernize Polish text. [ver. 1.0.0]
This is a long list of all submissions, if you want to see only the best, click leaderboard.
# | submitter | when | ver. | description | dev-0 CharMatch | test-A CharMatch | |
---|---|---|---|---|---|---|---|
2 | p/tlen | 2022-07-07 16:06 | 1.0.0 | Lucene Transducers ver. 0.27-SNAPSHOT extended=yes rule-based | 49.2 | 48.4 | |
1 | p/tlen | 2022-07-07 16:06 | 1.0.0 | Lucene Transducers ver. 0.27-SNAPSHOT extended=no rule-based | 49.1 | 48.4 | |
3 | p/tlen | 2022-07-07 15:40 | 1.0.0 | just copy the Tesseract output tesseract | 20.1 | 27.6 |