explosion/spaCy
1
1
Fork 0
mirror of https://github.com/explosion/spaCy.git synced 2025-06-30 01:43:21 +03:00
Code Issues Packages Projects Releases Wiki Activity

Fix test.

Browse Source
This commit is contained in:
Raphael Mitsch 2022-10-27 13:28:17 +02:00
parent ba91d0d1d9
commit ace5655fe1
1 changed files with 16 additions and 19 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

35
spacy/tests/pipeline/test_entity_linker.py
View File

@ -1217,39 +1217,35 @@ def test_nel_candidate_processing():
""" """
train_data = [ train_data = [
( (
"The sky over New York is blue.", "The sky is blue.",
{ {
"sent_starts": [1, 0, 0, 0, 0, 0, 0, 0], "sent_starts": [1, 0, 0, 0, 0],
}, },
), ),
( (
"They visited New York.", "They visited New York.",
{ {
"sent_starts": [1, 0, 0, 0, 0], "sent_starts": [1, 0, 0, 0, 0],
"entities": [(13, 21, "GPE")],
},
),
("", {}),
(
"New York is a city.",
{
"sent_starts": [1, 0, 0, 0, 0, 0],
"entities": [(0, 8, "GPE")],
}, },
), ),
# (
# "",
# {}
# ),
# (
# "New York is a city.",
# {
# "sent_starts": [1, 0, 0, 0, 0, 0],
# }
# ),
] ]
nlp = English() nlp = English()
# Add a custom rule-based component to mimick NER nlp.add_pipe("sentencizer")
ruler = nlp.add_pipe("entity_ruler", last=True)
ruler.add_patterns([{"label": "GPE", "pattern": [{"LOWER": "new york"}]}]) # type: ignore
vector_length = 3 vector_length = 3
train_examples = [] train_examples = []
for text, annotation in train_data: for text, annotation in train_data:
doc = nlp(text) train_examples.append(Example.from_dict(nlp(text), annotation))
train_examples.append(Example.from_dict(doc, annotation))
def create_kb(vocab): def create_kb(vocab):
# create artificial KB # create artificial KB
@ -1266,8 +1262,9 @@ def test_nel_candidate_processing():
losses = {} losses = {}
nlp.update(train_examples, sgd=optimizer, losses=losses) nlp.update(train_examples, sgd=optimizer, losses=losses)
# adding additional components that are required for the entity_linker # Add a custom rule-based component to mimick NER
nlp.add_pipe("sentencizer", first=True) ruler = nlp.add_pipe("entity_ruler", before="entity_linker")
ruler.add_patterns([{"label": "GPE", "pattern": [{"LOWER": "new york"}]}]) # type: ignore
# this will run the pipeline on the examples and shouldn't crash # this will run the pipeline on the examples and shouldn't crash
nlp.evaluate(train_examples) nlp.evaluate(train_examples)