Hello, I trained a model from scratch :
I generated .box .tiff files I generated lstmf files with .box and .tiff files for file in *.tiff; do echo $file base=`basename $file .tiff` tesseract --psm 7 --oem 1 $file $base ./tessdata/configs/lstm.train done I generated unicharset with unicharset_extractor unicharset_extractor *.tiff I generated .traineddata with combine_lang_model combine_lang_model \ --input_unicharset unicharset \ --script_dir ./tesseract/langdata_lstm-master \ --output_dir output \ --pass_through_recoder \ --lang_is_rtl \ --lang XXX Then i trained a new model with lstmtraining lstmtraining \ --traineddata XXX.traineddata \ --debug_interval -1 \ --net_spec '[1,40,0,1 Ct5,5,64 Mp3,3 Lfys128 Lbx256 Lbx256 O1c36]' \ --model_output ../output/ \ --train_listfile list.train \ --eval_listfile list.eval At the end of the training at iteration 3599 : Finished! Error rate = 0 I generated the model with the last checkpoint lstmtraining --stop_training \ --continue_from model0.086_544.checkpoint \ --traineddata XXX.traineddata \ --model_output model/XXX.traineddata I put XXX.traineddata to tessdata and with tesseract command -l XXX i have bad results. For example with plateN836.lstmf lstmeval give me : Truth:SM363XT OCR :SM363XT With with tesseract command -l XXX on plateN836.tiff i have HPYPPY9P44T. What's wrong ? I don't understand why i have different result with lstmeval and tesseract command. Thank you for your help. 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 tesseract-ocr+unsubscr...@googlegroups.com. To post to this group, send email to tesseract-ocr@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/1441cae2-c052-4c92-abf6-0d6c644817fc%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.