I seem to recall that Richard Zens implemented something like the future cost estimation of the distortion penalty in the workshop in 2006, so it may still be in the code, somewhere. (I also seem to recall that he found it didn't work, so it may not).
On Tue, Mar 6, 2012 at 1:59 PM, Lane Schwartz <[email protected]> wrote: > Moore & Quirk (2007), presented at MT Summit XI, describe a mechanism > for incorporating a distortion penalty estimate in the future cost > calculation, and also describe a type of early pruning. > > I was wondering if either or both of these techniques are currently > implemented in Moses. > > Thanks, > Lane Schwartz > _______________________________________________ > Moses-support mailing list > [email protected] > http://mailman.mit.edu/mailman/listinfo/moses-support _______________________________________________ Moses-support mailing list [email protected] http://mailman.mit.edu/mailman/listinfo/moses-support
