2018-07-24 21:38:44 +00:00
|
|
|
# coding: utf8
|
2017-08-18 20:27:42 +00:00
|
|
|
from __future__ import unicode_literals
|
2018-07-24 21:38:44 +00:00
|
|
|
|
2017-08-18 20:27:42 +00:00
|
|
|
import pytest
|
|
|
|
import numpy
|
2018-07-24 21:38:44 +00:00
|
|
|
from spacy.vocab import Vocab
|
|
|
|
from spacy.language import Language
|
|
|
|
from spacy.pipeline import DependencyParser
|
|
|
|
from spacy.syntax.arc_eager import ArcEager
|
|
|
|
from spacy.tokens import Doc
|
|
|
|
from spacy.syntax._beam_utils import ParserBeam
|
|
|
|
from spacy.syntax.stateclass import StateClass
|
|
|
|
from spacy.gold import GoldParse
|
2017-08-18 20:27:42 +00:00
|
|
|
|
|
|
|
|
|
|
|
@pytest.fixture
|
|
|
|
def vocab():
|
|
|
|
return Vocab()
|
|
|
|
|
2018-07-24 21:38:44 +00:00
|
|
|
|
2017-08-18 20:27:42 +00:00
|
|
|
@pytest.fixture
|
|
|
|
def moves(vocab):
|
|
|
|
aeager = ArcEager(vocab.strings, {})
|
|
|
|
aeager.add_action(2, 'nsubj')
|
|
|
|
aeager.add_action(3, 'dobj')
|
|
|
|
aeager.add_action(2, 'aux')
|
|
|
|
return aeager
|
|
|
|
|
|
|
|
|
|
|
|
@pytest.fixture
|
|
|
|
def docs(vocab):
|
|
|
|
return [Doc(vocab, words=['Rats', 'bite', 'things'])]
|
|
|
|
|
|
|
|
@pytest.fixture
|
|
|
|
def states(docs):
|
|
|
|
return [StateClass(doc) for doc in docs]
|
|
|
|
|
|
|
|
@pytest.fixture
|
|
|
|
def tokvecs(docs, vector_size):
|
|
|
|
output = []
|
|
|
|
for doc in docs:
|
|
|
|
vec = numpy.random.uniform(-0.1, 0.1, (len(doc), vector_size))
|
|
|
|
output.append(numpy.asarray(vec))
|
|
|
|
return output
|
|
|
|
|
|
|
|
|
|
|
|
@pytest.fixture
|
|
|
|
def golds(docs):
|
|
|
|
return [GoldParse(doc) for doc in docs]
|
|
|
|
|
|
|
|
|
|
|
|
@pytest.fixture
|
|
|
|
def batch_size(docs):
|
|
|
|
return len(docs)
|
|
|
|
|
|
|
|
|
|
|
|
@pytest.fixture
|
|
|
|
def beam_width():
|
|
|
|
return 4
|
|
|
|
|
|
|
|
|
|
|
|
@pytest.fixture
|
|
|
|
def vector_size():
|
|
|
|
return 6
|
|
|
|
|
|
|
|
|
|
|
|
@pytest.fixture
|
|
|
|
def beam(moves, states, golds, beam_width):
|
2017-08-19 02:15:46 +00:00
|
|
|
return ParserBeam(moves, states, golds, width=beam_width, density=0.0)
|
2017-08-18 20:27:42 +00:00
|
|
|
|
2018-07-24 21:38:44 +00:00
|
|
|
|
2017-08-18 20:27:42 +00:00
|
|
|
@pytest.fixture
|
|
|
|
def scores(moves, batch_size, beam_width):
|
|
|
|
return [
|
|
|
|
numpy.asarray(
|
|
|
|
numpy.random.uniform(-0.1, 0.1, (batch_size, moves.n_moves)),
|
|
|
|
dtype='f')
|
|
|
|
for _ in range(batch_size)]
|
|
|
|
|
|
|
|
|
|
|
|
def test_create_beam(beam):
|
|
|
|
pass
|
|
|
|
|
|
|
|
|
|
|
|
def test_beam_advance(beam, scores):
|
|
|
|
beam.advance(scores)
|
|
|
|
|
|
|
|
|
|
|
|
def test_beam_advance_too_few_scores(beam, scores):
|
|
|
|
with pytest.raises(IndexError):
|
|
|
|
beam.advance(scores[:-1])
|
2018-07-24 21:38:44 +00:00
|
|
|
|
|
|
|
|
|
|
|
def test_beam_parse():
|
|
|
|
nlp = Language()
|
|
|
|
nlp.add_pipe(DependencyParser(nlp.vocab), name='parser')
|
|
|
|
nlp.parser.add_label('nsubj')
|
|
|
|
nlp.parser.begin_training([], token_vector_width=8, hidden_width=8)
|
|
|
|
doc = nlp.make_doc('Australia is a country')
|
|
|
|
nlp.parser(doc, beam_width=2)
|