When I run lstmtraining as
/usr/bin/lstmtraining \
--debug_interval 0 \
--traineddata /usr/share/tesseract-ocr/4.00/tessdata/eng.traineddata \
--continue_from /output/eng.lstm \
--net_spec '[1,36,0,1 Ct3,3,16 Mp3,3 Lfys48 Lfx96 Lrx96 Lfx256 O1c111]'
\
--model_output /output/trained \
--learning_rate 20e-4 \
--train_listfile /input/train.files \
--eval_listfile /input/eval.files \
--max_iterations 5000
I'm only getting training error displayed in the stdout log eg:
At iteration 268/300/337, Mean rms=3.577%, delta=21.465%, char
train=87.195%, word train=81.204%, skip ratio=12.333%,
New best char error = 87.195 wrote checkpoint.
Is the eval_listfile actually used for anything? I'm hoping to
automatically perform model evaluation (ideally with early stopping) as
part of the training process (i.e. without calling lstmeval).
--
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/8e57e715-5537-4f8d-90d8-f2b96afddf6c%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
