I had used the new text2image program with the tesstrain.sh utility for generating the box/tiff pairs and trained on a large set of fonts -
see https://github.com/Shreeshrii/imagessan/blob/master/san95-langdata/language-specific.sh and https://github.com/Shreeshrii/imagessan/blob/master/san95-langdata/san.font_properties Additional unicharset entries are probably because different devanagari fonts have different number of glyphs for conjuncts (specially Siddhanta, Sanskrit2003, Chandas and Uttara). I do not currently have that setup (cygwin/msys2 with tesseract) to test. ShreeDevi ____________________________________________________________ भजन - कीर्तन - आरती @ http://bhajans.ramparivar.com On Wed, Sep 21, 2016 at 12:54 PM, RKVS Raman <[email protected]> wrote: > Hello Shridevi, > > Thanks for clarifying on the current status of cube. I has worked with > tesseract long back. So I will leave the cube module for the time being and > focus on using the old adaptive classifier. > > BTW I tried training https://github.com/Shreeshrii/imagessan/blob/ > master/san95-langdata/san.training_text on Noto Sans Devnagari and I > could get only 1165 entries in unicharset. > > How did you manage to get 1645 entries with it? > > > > Best Regards > -Raman > > > > On Wed, Sep 21, 2016 at 9:21 AM, ShreeDevi Kumar <[email protected]> > wrote: > >> For Sanskrit, please see https://github.com/Shreeshrii/imagessan >> >> where I have added the training sources as well as traineddata for two >> versions of training. In the testing I did on a small sample of images, it >> seemed to perform better than the 3.04 san.traineddata. >> >> You are welcome to try using them and provide feedback. >> >> ShreeDevi >> ____________________________________________________________ >> भजन - कीर्तन - आरती @ http://bhajans.ramparivar.com >> >> On Tue, Sep 20, 2016 at 7:05 PM, rkvsraman <[email protected]> wrote: >> >>> Hello, >>> >>> In tessdata , I see cube models only for hindi and not for Marathi and >>> Sanskrit thought they have the same script. >>> >>> Any particular reason for this? >>> >>> -- >>> 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 [email protected]. >>> To post to this group, send email to [email protected]. >>> Visit this group at https://groups.google.com/group/tesseract-ocr. >>> To view this discussion on the web visit https://groups.google.com/d/ms >>> gid/tesseract-ocr/83d5c408-c869-4c16-8847-78ba2d250763%40goo >>> glegroups.com >>> <https://groups.google.com/d/msgid/tesseract-ocr/83d5c408-c869-4c16-8847-78ba2d250763%40googlegroups.com?utm_medium=email&utm_source=footer> >>> . >>> For more options, visit https://groups.google.com/d/optout. >>> >> >> -- >> 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 [email protected]. >> To post to this group, send email to [email protected]. >> Visit this group at https://groups.google.com/group/tesseract-ocr. >> To view this discussion on the web visit https://groups.google.com/d/ms >> gid/tesseract-ocr/CAG2NduXv9SJy%3D9U3rvQAjyu1KVvM-GMiWCYc7F_ >> 0V0ZB7AQkhA%40mail.gmail.com >> <https://groups.google.com/d/msgid/tesseract-ocr/CAG2NduXv9SJy%3D9U3rvQAjyu1KVvM-GMiWCYc7F_0V0ZB7AQkhA%40mail.gmail.com?utm_medium=email&utm_source=footer> >> . >> >> For more options, visit https://groups.google.com/d/optout. >> > > -- > 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 [email protected]. > To post to this group, send email to [email protected]. > 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/CABFygUDpwgih4_YehFMNoF9B_WLRoN%2BiB6YfxQu6rSbszvXzfA% > 40mail.gmail.com > <https://groups.google.com/d/msgid/tesseract-ocr/CABFygUDpwgih4_YehFMNoF9B_WLRoN%2BiB6YfxQu6rSbszvXzfA%40mail.gmail.com?utm_medium=email&utm_source=footer> > . > > For more options, visit https://groups.google.com/d/optout. > -- 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 [email protected]. To post to this group, send email to [email protected]. 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/CAG2NduXcrUC%3DbRpwsZ9nPGHriLHNmvXBJYNzCJps1nW55w6WTw%40mail.gmail.com. For more options, visit https://groups.google.com/d/optout.
