Since it is binary phrase table with 'on-demand' loading, is that why it takes the same time irrespective of phrase table size?
Amit On Tue, Nov 23, 2010 at 12:23 PM, Amit Abbi <ada2...@columbia.edu> wrote: > Hi, > > When trying to decode using lattice input on an unfiltered binarized phrase > table(~ 692MB), decoding took the following times:- > > Single Path for each sentence ~ 1Hr 30 minutes > Mutiple paths for each sentence (~5000 on average) ~ 1 Hr 35 minutes. > > However, on using a filtered binarized phrase table (containing only the > relevant phrases - 14 times smaller than the original phrase table), the > decoder still took the same amount of time > > Mutiple paths for each sentence (~5000 on average) ~ 1 Hr 34 minutes. > > I understand that binarizing the phrase table causes the increase in time > as compared to the textual version. However, does the size of the binary > phrase table not matter? Are there any ways one can decrease the decoding > time when using lattice inputs? > > Amit >
_______________________________________________ Moses-support mailing list Moses-support@mit.edu http://mailman.mit.edu/mailman/listinfo/moses-support
