I was fine-tuning the ara.traineddata in tessdata_best and I have benchmarked the output model and tessdata_best model. I have found that the tessdata_best model did better than the output model, so is it a normal thing that my output model did worse? If not what am I doing wrong?
I am using Tesseract 5.3.3 and ubuntu wsl. This is the training line: ```make LANG_TYPE=RTL MODEL_NAME=Tesseract_F_JSTOR_TRDG_Shaden PSM=13 START_MODEL=ara TESSDATA=../tessdata_best FINETUNE_TYPE=Layer MAX_ITERATIONS=21000 RATIO_TRAIN=0.80 EPOCHS=3 training -j12``` -- 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/62b94864-34ee-4382-af67-8f93ff9f2b4dn%40googlegroups.com.
