Hi Mailing,
I am having trouble understanding what the recaser is doing exactly when evaluating a (dev) test set. 1) Why do we need to train a recaser? Is there some documentation about which arguments to give to train-recaser.perl 2) As I am still working with non-factored models, does anybody have a suggestion how to train the recaser differently? 3) Why is there yet another moses.ini file here. I thought at this stage we are finished training. 4) Finally and a little off topic, does anybody in our research field have word-based human-aligned alignment files for some test set in other languages than fr-en available? Thanks for any advise Regards, Sanne
_______________________________________________ Moses-support mailing list [email protected] http://mailman.mit.edu/mailman/listinfo/moses-support
