76
επεξεργασίες
Βικιθήκη:Bots: Διαφορά μεταξύ των αναθεωρήσεων
καμία σύνοψη επεξεργασίας
Geraki (συζήτηση | Συνεισφορά) |
Χωρίς σύνοψη επεξεργασίας |
||
|
# Which OCR software do you use, Tesseract? Do you train it?
Please make some more tests (100 edits?). - [[user:geraki|<span style="color:green;">geraki</span>]] <sup>[[user_talk:geraki|(συζήτηση)]]</sup> 15:44, 23 Φεβρουαρίου 2020 (UTC)
Καλημέρα [[user:geraki|Geraki]] !
thank you for your suggestions to improve the bot !
To answer your questions:
# The OCR was intented to be a starting point, better than no text or junk text. So I didn't intend to correct all the pages manually.
# I used tesseract, version 3.04, with languages grc and fra, I did not train it.
Following your remarks; I will:
* try to use only grc
* upgrade tesseract and tessdata to the latest version (4.1.1 for tesseract)
* proceed to more edits
In case it is necessary, I will train tesseract.
With these actions, I hope to be able to provide a better text. [[Χρήστης:Apameia|Apameia]] ([[Συζήτηση χρήστη:Apameia|συζήτηση]]) 08:33, 25 Φεβρουαρίου 2020 (UTC)
| |||