Results 538 comments of Amit Dovev

You can use this image for testing the new feature: https://github.com/DanBloomberg/leptonica/blob/a14036fa5f5ea971/prog/w91frag.jpg

@harinath141, Please try with the latest commit from the master branch.

@Shreeshrii, Was this fixed with my patch? Can we close it?

Thanks, Does it drop entire lines or just a few words in some lines?

[Levenshtein language:c++ stars:>50 -license:gpl](https://github.com/search?q=Levenshtein+language%3Ac%2B%2B+stars%3A%3E50+-license%3Agpl) Edit: Indeed, RapidFuzz seems to be the best option.

>* C/C++ At least in theory, we can use a tool written in Python/Perl or any other language and use unix sockets to communicate with it. The visual debugger already...

>* when you pass a validation set, you expect checkpoint and final error rate to refer to that – not the training error >* today you expect error to mean...

Regarding the second issue/task, BCER->CER. Until this issue is fixed in Tesseract itself, perhaps the issue could be mitigated in tesstrain with a Python/Perl script?

@bertsky, What @stweil suggested is quite close to what I wanted to know. Your local workaround seems similar. Is @stweil's/your workaround is 'good enough'? Tesseract's C++ developers pool is quite...

@wollmers, GitHub repos: https://github.com/mbethke/Text-Levenshtein-Flexible https://github.com/gitpan/Text-Levenshtein-Flexible