According #2481 <https://github.com/tesseract-ocr/tesseract/issues/2481>, the glyph metrics aren't used in LSTM training.
You may just skip the warnings. Pndaza於 2019年6月22日星期六 UTC+8下午8時03分53秒寫道： > > When running combine_lang_model, the following waring occurs. > > Warning: properties incomplete for index 6 = ိ > Warning: properties incomplete for index 11 = ် > Warning: properties incomplete for index 22 = ြ > Warning: properties incomplete for index 39 = ွ > Warning: properties incomplete for index 42 = ့ > Warning: properties incomplete for index 45 = ဲ > Warning: properties incomplete for index 57 = ှ > Warning: properties incomplete for index 58 = ံ > Warning: properties incomplete for index 64 = ီ > Warning: properties incomplete for index 84 = ဩ > > Warning: properties incomplete for index 105 = ဪ > > > I checks Myanmar.unicharset > <https://github.com/tesseract-ocr/langdata/blob/master/Myanmar.unicharset> .ြ > > ဩ and ဪ are missing. > I added index for ဩ and waring disappear. > others contain in Myanmar.unicharset > <https://github.com/tesseract-ocr/langdata/blob/master/Myanmar.unicharset> > . > > The following is comment from unicharset.h > <https://github.com/tesseract-ocr/tesseract/blob/8de022ab1cd291ffaf4d24bb4fa07b97edfca4a7/src/ccutil/unicharset.h> > // Returns true if any of the top/bottom/width/bearing/advance > ranges/stats is empty. > > ိ 0 58,59,255,255,211,215,0,0,0,0 Myanmar 44 17 44 ိ # ိ [102d ] >> ီ 0 58,59,255,255,211,215,0,0,0,0 Myanmar 45 17 45 ီ # ီ [102e ] >> ဲ 0 58,59,255,255,211,215,0,0,0,0 Myanmar 49 17 49 ဲ # ဲ [1032 ] >> ံ 0 58,59,255,255,211,215,0,0,0,0 Myanmar 50 17 50 ံ # ံ [1036 ] >> ့ 0 0,0,255,255,211,215,0,0,0,0 Myanmar 51 17 51 ့ # ့ [1037 ] > > > bearing and advance for those are zero, > i look at Devanagari.unicharset > > ं 0 62,76,194,242,81,178,0,27,0,77 Devanagari 2 17 2 ं # ं [902 ] > > > and I changed for ိ > ိ 0 58,59,255,255,211,215,1,27,1,27 Myanmar 44 17 44 ိ # ိ [102d ] > Nomore Warining. > > > - > > min_bearing, max_bearing: how far from the usual start position does > the leftmost part of the character begin. > - > > min_advance, max_advance: how far from the printer’s cell left do we > advance to begin the next character. > > I readed doc but I can't understand and calculate. > > pls help me. > > > > > > -- 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 view this discussion on the web visit https://groups.google.com/d/msgid/tesseract-ocr/eb782f25-9b27-4303-9f0b-96a3204fd49e%40googlegroups.com.
