Hi everybody I am creating curpus data to train. It is more than 9000pages (like 3 encyclopedias) of tif for "Traditional Arabic" font (will add other fonts later)
will use Tesseract 4.00 Alpha LSTM any hints will be useful. I took same config of recent ara.traineddata (12mb). set char_spacing to 0.7 will use this lstm paras --learning_rate 10e-5 --net_spec '[1,0,0,1 Ct5,5,16 Mp3,3 Lfys64 Lfx128 Lrx128 Lfx256 O1c1]' --net_mode 192 --perfect_sample_delay 19 as referenced Also, It is good to share your experience with tesseract ocr engine if anybody has custom trained data for Arabic please share it -- 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/6bef0f0a-d0ba-432a-b6da-3a91fbbcb90b%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
