*Hi all,* *I am trying to use an MT engine trained on the 800,000 line EuroParlament corpus in an online MT gisting application.*
*The process occupies around 23 Gb of RAM that is a bit too much in our settings. * *Since I don't know what is the text user will be translating, **the phrase table can not be prefiltered. However probably there are other tools and techniques that can be used to make the application more efficient in terms of RAM?* *Thank you beforehand,* *Maxim*
_______________________________________________ Moses-support mailing list Moses-support@mit.edu http://mailman.mit.edu/mailman/listinfo/moses-support