Rooba
/
spaCy
mirror of https://github.com/explosion/spaCy.git
1
1
Fork 0
Code Issues Packages Projects Releases Wiki Activity

* Fix oracle bug in NER. Now getting 77% F on ontonotes

This commit is contained in:
Matthew Honnibal 2015-03-11 02:25:08 -04:00
parent 903f196b3f
commit 1c843934be
1 changed files with 5 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
spacy/syntax/ner.pyx
View File

@ -41,7 +41,7 @@ cdef bint _entity_is_sunk(const State *s, Transition* golds) except -1:
if not entity_is_open(s): if not entity_is_open(s):
return False return False
cdef const Transition* gold = &golds[(s.i - 1) + s.ent.start] cdef const Transition* gold = &golds[s.ent.start]
if gold.move != BEGIN and gold.move != UNIT: if gold.move != BEGIN and gold.move != UNIT:
return True return True
elif gold.label != s.ent.label: elif gold.label != s.ent.label:
@ -107,6 +107,8 @@ cdef class BiluoPushDown(TransitionSystem):
for i in range(self.n_moves): for i in range(self.n_moves):
if self.c[i].move == move and self.c[i].label == label: if self.c[i].move == move and self.c[i].label == label:
return self.c[i] return self.c[i]
else:
raise StandardError(name)
cdef Transition init_transition(self, int clas, int move, int label) except *: cdef Transition init_transition(self, int clas, int move, int label) except *:
# TODO: Apparent Cython bug here when we try to use the Transition() # TODO: Apparent Cython bug here when we try to use the Transition()
@ -148,6 +150,7 @@ cdef int _get_cost(const Transition* self, const State* s, GoldParse gold) excep
gold.c_ner[s.i].label, next_act, is_sunk) gold.c_ner[s.i].label, next_act, is_sunk)
return not is_gold return not is_gold
cdef bint _is_gold(int act, int tag, int g_act, int g_tag, cdef bint _is_gold(int act, int tag, int g_act, int g_tag,
int next_act, bint is_sunk): int next_act, bint is_sunk):
if g_act == MISSING: if g_act == MISSING:
@ -227,6 +230,7 @@ cdef int _do_begin(const Transition* self, State* s) except -1:
s.ents_len += 1 s.ents_len += 1
s.ent.start = s.i s.ent.start = s.i
s.ent.label = self.label s.ent.label = self.label
s.ent.end = 0
s.sent[s.i].ent_iob = 3 s.sent[s.i].ent_iob = 3
s.sent[s.i].ent_type = self.label s.sent[s.i].ent_type = self.label
s.i += 1 s.i += 1