diff options
Diffstat (limited to 'gi/pf/reachability.cc')
| -rw-r--r-- | gi/pf/reachability.cc | 9 |
1 files changed, 5 insertions, 4 deletions
diff --git a/gi/pf/reachability.cc b/gi/pf/reachability.cc index 59bc6ace..c10000f2 100644 --- a/gi/pf/reachability.cc +++ b/gi/pf/reachability.cc @@ -12,7 +12,7 @@ struct SState { int prev_trg_covered; }; -void Reachability::ComputeReachability(int srclen, int trglen, int src_max_phrase_len, int trg_max_phrase_len) { +void Reachability::ComputeReachability(int srclen, int trglen, int src_max_phrase_len, int trg_max_phrase_len, double filter_ratio) { typedef boost::multi_array<vector<SState>, 2> array_type; array_type a(boost::extents[srclen + 1][trglen + 1]); a[0][0].push_back(SState()); @@ -30,9 +30,10 @@ void Reachability::ComputeReachability(int srclen, int trglen, int src_max_phras } } a[0][0].clear(); - //cerr << "Final cell contains " << a[srclen][trglen].size() << " back pointers\n"; - if (a[srclen][trglen].size() == 0) { - cerr << "Sentence with length (" << srclen << ',' << trglen << ") violates reachability constraints\n"; + //cerr << srclen << "," << trglen << ": Final cell contains " << a[srclen][trglen].size() << " back pointers\n"; + size_t min_allowed = (src_max_phrase_len + 1) * (trg_max_phrase_len + 1) * (filter_ratio * filter_ratio); + if (a[srclen][trglen].size() < min_allowed) { + cerr << "Sequence pair with lengths (" << srclen << ',' << trglen << ") violates reachability constraint of min indegree " << min_allowed << " with " << a[srclen][trglen].size() << " in edges\n"; return; } |