make trained model from checkpoint files

[z160896]

@zdenop I have come to the last hurdle! I can finish training the model successfully. I like to make the trained model from files in checkpoints dir from the instruction. can you help how to get it work. appreciated.

steps:

1. make traineddata MODEL_NAME=hwmodel OUTPUT_DIR=data/hwmodel
2. make training MODEL_NAME=hwmodel START_MODEL=eng LANG_TYPE=Eng
3. make traineddata CHECKPOINT_FILES="$(find data/hwmodel/checkpoints -name '*.checkpoint' -mtime -21)" MODEL_NAME=hwmodel OUTPUT_DIR=/mnt/d/git
   make: *** No rule to make target 'data/hwmodel/tessdata_best/hwmodel_19.699000_9449_9800.checkpoint
   data/hwmodel/tessdata_best/hwmodel_20.407000_7080_7300.checkpoint
   data/hwmodel/tessdata_best/hwmodel_21.067000_6516_6700.checkpoint
   data/hwmodel/tessdata_best/hwmodel_21.878000_6042_6200.checkpoint
   data/hwmodel/tessdata_best/hwmodel_22.593000_4598_4700.checkpoint
   data/hwmodel/tessdata_best/hwmodel_23.559000_4120_4200.checkpoint
   data/hwmodel/tessdata_best/hwmodel_24.341000_3831_3900.checkpoint
   data/hwmodel/tessdata_best/hwmodel_25.543000_2463_2500.checkpoint
   data/hwmodel/tessdata_best/hwmodel_26.484000_2271_2300.checkpoint
   data/hwmodel/tessdata_best/hwmodel_27.441000_1980_2000.checkpoint
   data/hwmodel/tessdata_best/hwmodel_28.992000_1683_1700.checkpoint
   data/hwmodel/tessdata_best/hwmodel_30.096000_1491_1500.checkpoint
   data/hwmodel/tessdata_best/hwmodel_31.523000_1193_1200.checkpoint
   data/hwmodel/tessdata_best/hwmodel_32.787000_1093_1100.checkpoint
   data/hwmodel/tessdata_best/hwmodel_35.179000_796_800.checkpoint
   data/hwmodel/tessdata_best/hwmodel_36.494000_597_600.checkpoint
   data/hwmodel/tessdata_best/hwmodel_38.070000_399_400.checkpoint
   data/hwmodel/tessdata_best/hwmodel_39.380000_300_300.checkpoint
   data/hwmodel/tessdata_best/hwmodel_42.453000_200_200.checkpoint
   data/hwmodel/tessdata_best/hwmodel_45.193000_100_100.traineddata', needed by 'traineddata'. Stop.

[stale]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.

[bertsky]

@z160896 to convert a checkpoint to a traineddata file, just do:

lstmtraining --stop_training --continue_from path/to/checkpoint --traineddata path/to/proto.traineddata --model_output path/to/trainedata

In your case, that should be:

lstmtraining --stop_training --continue_from data/hwmodel/tessdata_best/hwmodel_19.699000_9449_9800.checkpoint --traineddata data/hwmodel/hwmodel.traineddata --model_output data/hwmodel.traineddata