Hi, yes, they use the same decoding algorithm and phrase table format. The only difference is the additional annotation of source and target non-terminals in syntax models.
-phi 2012/2/11 dongxinghua0213 <[email protected]>: > hi, > all, Is Hierarchical phrase model and Syntactic models in the same > framework(in moses toolkit)? Are they organized in the same data structure > ? e,g, in prefix tree structure, > of -> the -> JJ/ADJ -> problem -> of -> NP/NP-GEN -> des ADJ_1 Problems > NP-GEN_2 > > in a span ,we search the tree to see if each node matchs the span, in node > JJ , we have to chech the smaller span to see if target side nonterminal is > ADJ. Is the case in Hierarchical phrase model in which there is only a > nonterminal X ? If it is true , does it slow the Hierarchical phrase > decoder ? > > Thank you for reply ! Best wishes ! > > _______________________________________________ Moses-support mailing list [email protected] http://mailman.mit.edu/mailman/listinfo/moses-support
