OS: Windows
$ make training TESSDATA=/c/Users/gahse/Desktop/Projects/lib/dup/tesstrain/data/tessdata MODEL_NAME=jpn_font START_MODEL=jpn
You are using make version: 4.4.1
tesseract "data/jpn_font-ground-truth/001pg1_1.tif" data/jpn_font-ground-truth/001pg1_1 --psm 13 lstm.train
read_params_file: Can't open lstm.train
tesseract "data/jpn_font-ground-truth/001pg1_2.tif" data/jpn_font-ground-truth/001pg1_2 --psm 13 lstm.train
read_params_file: Can't open lstm.train
tesseract "data/jpn_font-ground-truth/001pg1_3.tif" data/jpn_font-ground-truth/001pg1_3 --psm 13 lstm.train
read_params_file: Can't open lstm.train
python shuffle.py 0 "data/jpn_font/all-lstmf"
python generate_eval_train.py data/jpn_font/all-lstmf 0.90
lstmtraining \
--debug_interval 0 \
--traineddata data/jpn_font/jpn_font.traineddata \
--old_traineddata C:/Users/gahse/Desktop/Projects/lib/dup/tesstrain/data/tessdata/jpn.traineddata \
--continue_from data/jpn/jpn_font.lstm \
--learning_rate 0.0001 \
--model_output data/jpn_font/checkpoints/jpn_font \
--train_listfile data/jpn_font/list.train \
--eval_listfile data/jpn_font/list.eval \
--max_iterations 10000 \
--target_error_rate 0.01 \
2>&1 | tee -a data/jpn_font/training.log
Loaded file data/jpn/jpn_font.lstm, unpacking...
Failed to continue from: data/jpn/jpn_font.lstm
lstmtraining \
--stop_training \
--continue_from data/jpn_font/checkpoints/jpn_font_checkpoint \
--traineddata data/jpn_font/jpn_font.traineddata \
--model_output data/jpn_font.traineddata
Failed to read continue from: data/jpn_font/checkpoints/jpn_font_checkpoint
make: *** [Makefile:325: data/jpn_font.traineddata] Error 1
Hi im having issue
1. What is the cause that failed from .lstm?
2. The checkpoint file will auto generated or need to create manually? What is the format?