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

Modernise BILUO tests

This commit is contained in:
Ines Montani 2017-01-12 23:39:18 +01:00
parent 09acfbca01
commit 5e4f5ebfc8
1 changed files with 11 additions and 16 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

27
spacy/tests/gold/test_biluo.py
View File

@ -1,48 +1,43 @@
# coding: utf-8
from __future__ import unicode_literals from __future__ import unicode_literals
from ...gold import biluo_tags_from_offsets from ...gold import biluo_tags_from_offsets
from ...vocab import Vocab
from ...tokens.doc import Doc from ...tokens.doc import Doc
import pytest import pytest
@pytest.fixture def test_gold_biluo_U(en_vocab):
def vocab(): orths_and_spaces = [('I', True), ('flew', True), ('to', True),
return Vocab() ('London', False), ('.', True)]
doc = Doc(en_vocab, orths_and_spaces=orths_and_spaces)
def test_U(vocab):
orths_and_spaces = [('I', True), ('flew', True), ('to', True), ('London', False),
('.', True)]
doc = Doc(vocab, orths_and_spaces=orths_and_spaces)
entities = [(len("I flew to "), len("I flew to London"), 'LOC')] entities = [(len("I flew to "), len("I flew to London"), 'LOC')]
tags = biluo_tags_from_offsets(doc, entities) tags = biluo_tags_from_offsets(doc, entities)
assert tags == ['O', 'O', 'O', 'U-LOC', 'O'] assert tags == ['O', 'O', 'O', 'U-LOC', 'O']
def test_BL(vocab): def test_gold_biluo_BL(en_vocab):
orths_and_spaces = [('I', True), ('flew', True), ('to', True), ('San', True), orths_and_spaces = [('I', True), ('flew', True), ('to', True), ('San', True),
('Francisco', False), ('.', True)] ('Francisco', False), ('.', True)]
doc = Doc(vocab, orths_and_spaces=orths_and_spaces) doc = Doc(en_vocab, orths_and_spaces=orths_and_spaces)
entities = [(len("I flew to "), len("I flew to San Francisco"), 'LOC')] entities = [(len("I flew to "), len("I flew to San Francisco"), 'LOC')]
tags = biluo_tags_from_offsets(doc, entities) tags = biluo_tags_from_offsets(doc, entities)
assert tags == ['O', 'O', 'O', 'B-LOC', 'L-LOC', 'O'] assert tags == ['O', 'O', 'O', 'B-LOC', 'L-LOC', 'O']
def test_BIL(vocab): def test_gold_biluo_BIL(en_vocab):
orths_and_spaces = [('I', True), ('flew', True), ('to', True), ('San', True), orths_and_spaces = [('I', True), ('flew', True), ('to', True), ('San', True),
('Francisco', True), ('Valley', False), ('.', True)] ('Francisco', True), ('Valley', False), ('.', True)]
doc = Doc(vocab, orths_and_spaces=orths_and_spaces) doc = Doc(en_vocab, orths_and_spaces=orths_and_spaces)
entities = [(len("I flew to "), len("I flew to San Francisco Valley"), 'LOC')] entities = [(len("I flew to "), len("I flew to San Francisco Valley"), 'LOC')]
tags = biluo_tags_from_offsets(doc, entities) tags = biluo_tags_from_offsets(doc, entities)
assert tags == ['O', 'O', 'O', 'B-LOC', 'I-LOC', 'L-LOC', 'O'] assert tags == ['O', 'O', 'O', 'B-LOC', 'I-LOC', 'L-LOC', 'O']
def test_misalign(vocab): def test_gold_biluo_misalign(en_vocab):
orths_and_spaces = [('I', True), ('flew', True), ('to', True), ('San', True), orths_and_spaces = [('I', True), ('flew', True), ('to', True), ('San', True),
('Francisco', True), ('Valley.', False)] ('Francisco', True), ('Valley.', False)]
doc = Doc(vocab, orths_and_spaces=orths_and_spaces) doc = Doc(en_vocab, orths_and_spaces=orths_and_spaces)
entities = [(len("I flew to "), len("I flew to San Francisco Valley"), 'LOC')] entities = [(len("I flew to "), len("I flew to San Francisco Valley"), 'LOC')]
tags = biluo_tags_from_offsets(doc, entities) tags = biluo_tags_from_offsets(doc, entities)
assert tags == ['O', 'O', 'O', '-', '-', '-'] assert tags == ['O', 'O', 'O', '-', '-', '-']