explosion/spaCy
1
1
Fork 0
mirror of https://github.com/explosion/spaCy.git synced 2025-08-04 04:10:20 +03:00
Code Issues Packages Projects Releases Wiki Activity

Fix offset bug in set_annotations

Browse Source 
* Ignore labels in span finder scorer
This commit is contained in:
Adriane Boyd 2023-06-02 19:43:03 +02:00
parent f84b59d68a
commit bb62ee9450
2 changed files with 3 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
spacy/pipeline/span_finder.py
View File

@ -111,6 +111,7 @@ def span_finder_score(examples: Iterable[Example], **kwargs) -> Dict[str, Any]:
)
kwargs.setdefault("has_annotation", lambda doc: key in doc.spans)
kwargs.setdefault("allow_overlap", True)
kwargs.setdefault("labeled", False)
scores = Scorer.score_spans(examples, **kwargs)
scores.pop(f"{kwargs['attr']}_per_type", None)
return scores
@ -215,6 +216,7 @@ class SpanFinder(TrainablePipe):
break
elif self.min_length <= span_length:
doc.spans[self.spans_key].append(doc[start : end + 1])
offset += len(doc)
def update(
self,

2
spacy/tests/pipeline/test_span_finder.py
View File

@ -234,7 +234,7 @@ def test_overfitting_IO():
scores = nlp.evaluate(train_examples)
assert f"span_finder_{span_finder.spans_key}_f" in scores
# It's not perfect 1.0 F1 because it's designed to overgenerate for now.
assert scores[f"span_finder_{span_finder.spans_key}_f"] == 0.4
assert scores[f"span_finder_{span_finder.spans_key}_p"] == 0.75
assert scores[f"span_finder_{span_finder.spans_key}_r"] == 1.0
# also test that the spancat works for just a single entity in a sentence