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

* Fix morphology loading

This commit is contained in:
Matthew Honnibal 2015-09-10 14:52:23 +02:00
parent 7c660c5efc
commit 85c3fec1d1
1 changed files with 6 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

9
spacy/morphology.pyx
View File

@ -34,7 +34,6 @@ cdef class Morphology:
try: try:
tag_id = self.reverse_index[self.strings[tag]] tag_id = self.reverse_index[self.strings[tag]]
except KeyError: except KeyError:
print tag
raise raise
else: else:
tag_id = tag tag_id = tag
@ -61,12 +60,16 @@ cdef class Morphology:
cdef dict props cdef dict props
cdef int lemma cdef int lemma
cdef attr_t orth cdef attr_t orth
cdef attr_t tag_id
cdef int pos cdef int pos
cdef RichTagC rich_tag
for tag_str, entries in exc.items(): for tag_str, entries in exc.items():
tag = self.strings[tag_str] tag = self.strings[tag_str]
rich_tag = self.rich_tags[self.reverse_index[tag]] tag_id = self.reverse_index[tag]
rich_tag = self.rich_tags[tag_id]
for form_str, props in entries.items(): for form_str, props in entries.items():
cached = <MorphAnalysisC*>self.mem.alloc(1, sizeof(MorphAnalysisC)) cached = <MorphAnalysisC*>self.mem.alloc(1, sizeof(MorphAnalysisC))
cached.tag = rich_tag
orth = self.strings[form_str] orth = self.strings[form_str]
for name_str, value_str in props.items(): for name_str, value_str in props.items():
if name_str == 'L': if name_str == 'L':
@ -75,7 +78,7 @@ cdef class Morphology:
self.assign_feature(&cached.tag.morph, name_str, value_str) self.assign_feature(&cached.tag.morph, name_str, value_str)
if cached.lemma == 0: if cached.lemma == 0:
cached.lemma = self.lemmatize(rich_tag.pos, orth) cached.lemma = self.lemmatize(rich_tag.pos, orth)
self._cache.set(rich_tag.pos, orth, <void*>cached) self._cache.set(tag_id, orth, <void*>cached)
def lemmatize(self, const univ_pos_t pos, attr_t orth): def lemmatize(self, const univ_pos_t pos, attr_t orth):
if self.lemmatizer is None: if self.lemmatizer is None: