Hi all,
Can any one brief me regarding Tesseract 4 lstm training model from scratch ? I am done with LSTM fine tunning. Want to know what is the exact differrence in training from Scrach and in fine tunning ?
I followed following process:
1. tesstrain.sh to create starter data
2. lstmtraining for fine tuning with extracted model from best Trained data
3. lstmtraning --stop_training to generate .traineddata file