I followed the [documentation ](https://github.com/tesseract-ocr/tesstrain). I prepared 350 one line Arabic images (xx.png) and their transcript (xx.gt.txt) and starting training with START_MODEL=ara.

tesseract-ocr / tesstrain

Train Tesseract LSTM with make

Apache License 2.0

604 stars 178 forks source link

I followed the [documentation ](https://github.com/tesseract-ocr/tesstrain). I prepared 350 one line Arabic images (xx.png) and their transcript (xx.gt.txt) and starting training with START_MODEL=ara. #262

Closed soufieneghribi closed 3 years ago

soufieneghribi commented 3 years ago

I followed the documentation . I prepared 350 one line Arabic images (xx.png) and their transcript (xx.gt.txt) and starting training with START_MODEL=ara.

make training MODEL_NAME=elda START_MODEL=ara TESSDATA=data/ara_best

I am getting 100% as error rate.

I there something I am missing?

Originally posted by @soufieneghribi in https://github.com/tesseract-ocr/tesstrain/issues/252#issuecomment-854469078

soufieneghribi commented 3 years ago

This was fixed by LANG_TYPE=RTL. I'm closing this issue