* Load lemmatizer from disk in Vocab.from_dir

This commit is contained in:
Matthew Honnibal 2015-09-10 14:49:10 +02:00
parent 07686470a9
commit 90da3a695d
1 changed files with 20 additions and 15 deletions

View File

@ -38,19 +38,6 @@ EMPTY_LEXEME.repvec = EMPTY_VEC
cdef class Vocab: cdef class Vocab:
'''A map container for a language's LexemeC structs. '''A map container for a language's LexemeC structs.
''' '''
def __init__(self, get_lex_attr=None, tag_map=None, vectors=None):
if tag_map is None:
tag_map = {}
self.mem = Pool()
self._by_hash = PreshMap()
self._by_orth = PreshMap()
self.strings = StringStore()
self.get_lex_attr = get_lex_attr
self.morphology = Morphology(self.strings, tag_map, Lemmatizer({}, {}, {}))
self.length = 1
self._serializer = None
@classmethod @classmethod
def from_dir(cls, data_dir, get_lex_attr=None, vectors=None): def from_dir(cls, data_dir, get_lex_attr=None, vectors=None):
if not path.exists(data_dir): if not path.exists(data_dir):
@ -59,13 +46,31 @@ cdef class Vocab:
raise IOError("Path %s is a file, not a dir -- cannot load Vocab." % data_dir) raise IOError("Path %s is a file, not a dir -- cannot load Vocab." % data_dir)
tag_map = json.load(open(path.join(data_dir, 'tag_map.json'))) tag_map = json.load(open(path.join(data_dir, 'tag_map.json')))
cdef Vocab self = cls(get_lex_attr=get_lex_attr, vectors=vectors, tag_map=tag_map) lemmatizer = Lemmatizer.from_dir(path.join(data_dir, '..'))
cdef Vocab self = cls(get_lex_attr=get_lex_attr, vectors=vectors, tag_map=tag_map,
lemmatizer=lemmatizer)
self.load_lexemes(path.join(data_dir, 'strings.txt'), path.join(data_dir, 'lexemes.bin')) self.load_lexemes(path.join(data_dir, 'strings.txt'), path.join(data_dir, 'lexemes.bin'))
if vectors is None and path.exists(path.join(data_dir, 'vec.bin')): if vectors is None and path.exists(path.join(data_dir, 'vec.bin')):
self.repvec_length = self.load_rep_vectors(path.join(data_dir, 'vec.bin')) self.repvec_length = self.load_rep_vectors(path.join(data_dir, 'vec.bin'))
return self return self
def __init__(self, get_lex_attr=None, tag_map=None, vectors=None, lemmatizer=None):
if tag_map is None:
tag_map = {}
if lemmatizer is None:
lemmatizer = Lemmatizer({}, {}, {})
self.mem = Pool()
self._by_hash = PreshMap()
self._by_orth = PreshMap()
self.strings = StringStore()
self.get_lex_attr = get_lex_attr
self.morphology = Morphology(self.strings, tag_map, lemmatizer)
self.length = 1
self._serializer = None
property serializer: property serializer:
def __get__(self): def __get__(self):
if self._serializer is None: if self._serializer is None:
@ -199,7 +204,7 @@ cdef class Vocab:
lexeme = <LexemeC*>addr lexeme = <LexemeC*>addr
fp.write_from(&lexeme.orth, sizeof(lexeme.orth), 1) fp.write_from(&lexeme.orth, sizeof(lexeme.orth), 1)
fp.write_from(&lexeme.flags, sizeof(lexeme.flags), 1) fp.write_from(&lexeme.flags, sizeof(lexeme.flags), 1)
fp.write_from(&lexeme.id, sizeof(lexeme.flags), 1) fp.write_from(&lexeme.id, sizeof(lexeme.id), 1)
fp.write_from(&lexeme.length, sizeof(lexeme.length), 1) fp.write_from(&lexeme.length, sizeof(lexeme.length), 1)
fp.write_from(&lexeme.orth, sizeof(lexeme.orth), 1) fp.write_from(&lexeme.orth, sizeof(lexeme.orth), 1)
fp.write_from(&lexeme.lower, sizeof(lexeme.lower), 1) fp.write_from(&lexeme.lower, sizeof(lexeme.lower), 1)