Hello! For fine tuning, is there a way to use eval_listfile to monitor the model's performance on evaluation fonts? I tried to pass this parameter to lstmtraining, but the log does not show any message related to using eval_listfile.
If eval_listfile has no effect on training, other than experimenting different models on the evaluation fonts with different target_error_rate or max_iterations, or the various checkpoint files saved from training, is there a more systematic way that people have been following to prevent overfitting? I'm following the instruction under "Fine Tuning for Impact" and use the following command (except for the last line): training/lstmtraining --model_output /path/to/output \ --continue_from /path/to/existing/model \ --traineddata /path/to/original/traineddata \ --target_error_rate 0.01 \ --train_listfile /path/to/list/of/filenames.txt \ --eval_listfile /path/to/eval_list/of/filenames.txt Thanks, Joan -- 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 post to this group, send email to [email protected]. Visit this group at https://groups.google.com/group/tesseract-ocr. To view this discussion on the web visit https://groups.google.com/d/msgid/tesseract-ocr/25070e70-b7d1-45f0-a4d8-d3b4fa7e213a%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
