Hi Hieu, Yes, that was it, we hadn't updated train-model.perl.
This latest change also runs fine for me (for Hiero training). Thanks!! Cheers, Alex On Mon, Jul 2, 2012 at 3:15 PM, Hieu Hoang <[email protected]> wrote: > i think it should work with factors now. > https://github.com/moses-smt/mosesdecoder/commit/dea9e778a256e99086d77d764e2a88336f96f7ab > > On 02/07/2012 18:29, Hieu Hoang wrote: >> hi alex >> >> i works ok on my end but the train-model.perl had to be updated as well. >> >> however, just noticed a problem when using it with factors. Are you >> using factors? >> >> On 02/07/2012 18:19, Alexander Fraser wrote: >>> Hi -- >>> >>> John Morgan and I found a bug in experiment.perl - the "ptImpl" is >>> being incorrectly passed as part of the phrase table filename in >>> filtering for tuning hierarchical models. Anyone have a quick fix for >>> this by chance? >>> >>> Thanks, Alex >>> _______________________________________________ >>> 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 _______________________________________________ Moses-support mailing list [email protected] http://mailman.mit.edu/mailman/listinfo/moses-support
