explosion/spaCy
1
1
Fork 0
mirror of https://github.com/explosion/spaCy.git synced 2024-12-26 18:06:29 +03:00
Code Issues Packages Projects Releases Wiki Activity

Add .blank() method to Parser. Start housing default dep labels and entity types within the Defaults class.

Browse Source
This commit is contained in:
Matthew Honnibal 2016-09-26 11:57:54 +02:00
parent 722199acb8
commit 2debc4e0a2
3 changed files with 18 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

13
spacy/language.py
View File

@ -29,6 +29,7 @@ from . import util
from .lemmatizer import Lemmatizer from .lemmatizer import Lemmatizer
from .attrs import TAG, DEP, ENT_IOB, ENT_TYPE, HEAD, PROB, LANG, IS_STOP from .attrs import TAG, DEP, ENT_IOB, ENT_TYPE, HEAD, PROB, LANG, IS_STOP
from .syntax.parser import get_templates
class BaseDefaults(object): class BaseDefaults(object):
@ -98,14 +99,14 @@ class BaseDefaults(object):
return Parser.load(self.path / 'deps', vocab, ArcEager) return Parser.load(self.path / 'deps', vocab, ArcEager)
else: else:
return Parser.blank(vocab, ArcEager, return Parser.blank(vocab, ArcEager,
Parser.default_templates('%s-parser' % self.lang)) features=self.parser_features, labels=self.parser_labels)
def Entity(self, vocab): def Entity(self, vocab):
if self.path and (self.path / 'ner').exists(): if self.path and (self.path / 'ner').exists():
return Parser.load(self.path / 'ner', vocab, BiluoPushDown) return Parser.load(self.path / 'ner', vocab, BiluoPushDown)
else: else:
return Parser.blank(vocab, BiluoPushdown, return Parser.blank(vocab, BiluoPushdown,
Parser.default_templates('%s-entity' % self.lang)) features=self.entity_features, labels=self.entity_labels)
def Matcher(self, vocab): def Matcher(self, vocab):
if self.path: if self.path:
@ -120,9 +121,13 @@ class BaseDefaults(object):
nlp.parser, nlp.parser,
nlp.entity] nlp.entity]
dep_labels = {0: {'ROOT': True}} parser_labels = {0: {'ROOT': True}}
ner_labels = {0: {'PER': True, 'LOC': True, 'ORG': True, 'MISC': True}} entity_labels = {0: {'PER': True, 'LOC': True, 'ORG': True, 'MISC': True}}
parser_features = get_templates('parser')
entity_features = get_templates('ner')
stop_words = set() stop_words = set()

8
spacy/syntax/parser.pyx
View File

@ -89,6 +89,14 @@ cdef class Parser:
model.load(str(path / 'model')) model.load(str(path / 'model'))
return cls(vocab, moves, model, **cfg) return cls(vocab, moves, model, **cfg)
@classmethod
def blank(cls, Vocab vocab, moves_class, **cfg):
moves = moves_class(vocab.strings, cfg.get('labels', {}))
templates = get_templates(cfg.get('features', tuple()))
model = ParserModel(templates)
return cls(vocab, moves, model, **cfg)
def __init__(self, Vocab vocab, transition_system, ParserModel model, **cfg): def __init__(self, Vocab vocab, transition_system, ParserModel model, **cfg):
self.moves = transition_system self.moves = transition_system
self.model = model self.model = model

2
spacy/tests/conftest.py
View File

@ -3,6 +3,7 @@ import os
import spacy import spacy
@pytest.fixture(scope="session") @pytest.fixture(scope="session")
def EN(): def EN():
return spacy.load("en") return spacy.load("en")
@ -21,7 +22,6 @@ def pytest_addoption(parser):
help="include slow tests") help="include slow tests")
def pytest_runtest_setup(item): def pytest_runtest_setup(item):
for opt in ['models', 'vectors', 'slow']: for opt in ['models', 'vectors', 'slow']:
if opt in item.keywords and not item.config.getoption("--%s" % opt): if opt in item.keywords and not item.config.getoption("--%s" % opt):