Hi All,
I am trying to train the language of right to left writing sequence. My
training procedure is as follows:
1. I have created the images of the format .png along with the respective
transcription in .txt file of utf-8 format
e-g
oo1.png (image is in gray scale)
001.txt
2. Create a .txt file names list and write the following
001.png oo1.txt
3. Then I write the following command (taken from the wiki) to train the
data
OCROSCRIPTS=$PWD/scripts ./ocroscript train-bpnet-isolated --degrade=0 --
nhidden=15 --epochs=5 --learningrate=0.2 --testportion=0.2 --normalize=1
--shuffle=1 ../Isolated Chracters/list bpnet.classifier
After running the above command the following message arise
Usage: ocroscript train-bpnet-isolated [options] <dataset> <output>
<dataset> - a list of "image_path text_path" lines
Isolated training: every line should contain one character only
Options:
--nhidden=... - the number of hidden units
--epochs=... - the number of epochs to train
--learningrate=... - weights change coefficient
--testportion=... - how much data to leave for validation
--normalize=... - normalize means and dispersions of input
--shuffle=... - randomize the order of data
--degrade=... - the number of passes with degradation
(set to 0 to make one pass on pure data)
can anybody tell me what is missing in the procedure .
Regards
Ainie
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"ocropus" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at
http://groups.google.com/group/ocropus?hl=en
-~----------~----~----~----~------~----~------~--~---
