+ @theraysmith Two samples are certainly not enough for LSTM training. You have not specified whether you are training for a specific font for this particular numeric field of 5 digits. In case you are planning to use this traineddata ONLY for this numbers field, I think, you should train with all 100,000 combinations, since LSTM prefers large number of samples of the kind it is going to encounter.
cc:ing Ray, for confirmation. ShreeDevi ____________________________________________________________ भजन - कीर्तन - आरती @ http://bhajans.ramparivar.com On Sun, Jan 29, 2017 at 5:20 PM, Essam Zaky <[email protected]> wrote: > Dears > > I have question > > is stlm can recognize not seen sequence of characters ? > > > here it's example for what i mean > > if I have filed contain number consists of five digits > > when i train the LSTM should i generate all combinations of digits and > feed it to LSTM trainer or just i can give few samples cover all digits > > all combinations like the following > 00000 > 00001 > 00010 > ... > ... > ... > 99999 > > > and here there are few samples contain all digits > 01234 > 56789 > are the two samples enough or i must generate the all combinations as > explained in the previous step > > > Best Regards > > > > -- > 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/f997d921-97f1-4896-9725-b71331def675% > 40googlegroups.com > <https://groups.google.com/d/msgid/tesseract-ocr/f997d921-97f1-4896-9725-b71331def675%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/msgid/tesseract-ocr/CAG2NduWzqiQ990gcBc8OnqZ8JeEfL3s7Wg1RteoeWDXQQkbT0A%40mail.gmail.com. For more options, visit https://groups.google.com/d/optout.
