> By checking the "Training" instruction page and seeing what the
> current code can do, the farthest I got was "retrain=1 ocropus
> trainseg <cmodel> <bookdir>", but I got the impression it just
> replaced the old cmodel.

"retrain" just changes the naming convention used by

If you want to continue training an existing model, you need to use
linerec_cpreload; this is used for book adaptation, for example.  It
works quite well (see the publications).

> That language modeling might be another story, but cmodel training
> would be a start.

For adding words to the language model, you need OpenFST in general
(since you want to use operations like minimization).   You can code
that either in C++ or using a Python script using the pyopenfst
bindings (project on Google Code).

Tom

--~--~---------~--~----~------------~-------~--~----~
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
-~----------~----~----~----~------~----~------~--~---

Reply via email to