Hi all,
I have run decoder on both Window XP SP 2 and Fedora C7. I used the same
training model (phrase-table, language model, reordering, ...) when run
decoder on each OS. But the results I received is different from each other.
Ex : I run moses with input sentence : "I love you". The output sentence I
received :
- On Window : "Tôi tình yêu cô"
- On Fedora : "Tôi yêu cô"
The translation sentence quality received when running decoder on Fedora is
better than does on Window.
Why there is different between running decoder on Window and Fedora?
Could you help me to explain about this problem!
Best Regards,
_______________________________________________
Moses-support mailing list
[email protected]
http://mailman.mit.edu/mailman/listinfo/moses-support
