oem Tesseract + lstm

347 views
Skip to first unread message

Ibr

unread,
Oct 29, 2017, 10:44:00 AM10/29/17
to tesseract-ocr
Hi

I'm using: tesseract 4.00.00dev-692-gad5ee18 and leptonica-1.74.4 I want to use the oem 2, which is "2    Tesseract + LSTM." for English language that means I need two traineddata, the traineddata with LSTM which is integrated with tesseract 4, and the traineddata which doesn't contain LSTM, right?

my question is, for the argument "--tessdata-dir PATH" how can I specify to paths for both traineddata files? or should I put both traineddata files in the same path? in this case their name cant be identical, should I name each one in a certain name in order the tesseract will recognize them both when they are in the same directory?

Thank you

ShreeDevi Kumar

unread,
Oct 29, 2017, 11:34:06 AM10/29/17
to tesser...@googlegroups.com
The same traineddata file should have  files for both engines - legacy tesseract and lstm.

ShreeDevi
____________________________________________________________
भजन - कीर्तन - आरती @ http://bhajans.ramparivar.com

--
You received this message because you are subscribed to the Google Groups "tesseract-ocr" group.
To unsubscribe from this group and stop receiving emails from it, send an email to tesseract-ocr+unsubscribe@googlegroups.com.
To post to this group, send email to tesser...@googlegroups.com.
Visit this group at https://groups.google.com/group/tesseract-ocr.
To view this discussion on the web visit https://groups.google.com/d/msgid/tesseract-ocr/d1b3f967-0089-4444-979e-8cf8202588cb%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Ibr

unread,
Oct 30, 2017, 9:07:36 AM10/30/17
to tesseract-ocr
thank you for the answer.

how to do that, I mean have both traineddata in the same file? is that made by getting the legacy "old traineddata", unpack it, add LSTM to its contents then combine all of the components together?
or there is another way?


On Sunday, October 29, 2017 at 5:34:06 PM UTC+2, shree wrote:
The same traineddata file should have  files for both engines - legacy tesseract and lstm.

ShreeDevi
____________________________________________________________
भजन - कीर्तन - आरती @ http://bhajans.ramparivar.com

On Sun, Oct 29, 2017 at 8:14 PM, Ibr <ibr.h...@gmail.com> wrote:
Hi

I'm using: tesseract 4.00.00dev-692-gad5ee18 and leptonica-1.74.4 I want to use the oem 2, which is "2    Tesseract + LSTM." for English language that means I need two traineddata, the traineddata with LSTM which is integrated with tesseract 4, and the traineddata which doesn't contain LSTM, right?

my question is, for the argument "--tessdata-dir PATH" how can I specify to paths for both traineddata files? or should I put both traineddata files in the same path? in this case their name cant be identical, should I name each one in a certain name in order the tesseract will recognize them both when they are in the same directory?

Thank you

--
You received this message because you are subscribed to the Google Groups "tesseract-ocr" group.
To unsubscribe from this group and stop receiving emails from it, send an email to tesseract-oc...@googlegroups.com.
Reply all
Reply to author
Forward
0 new messages