explosion/spaCy
1
1
Fork 0
mirror of https://github.com/explosion/spaCy.git synced 2025-02-11 00:50:33 +03:00
Code Issues Packages Projects Releases Wiki Activity

Improve handling of missing values in NER

Browse Source
This commit is contained in:
Matthew Honnibal 2020-06-25 16:26:44 +02:00
parent b8c85e593b
commit 6bda23ad26
1 changed files with 28 additions and 24 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
spacy/gold/iob_utils.py
View File

@ -91,6 +91,10 @@ def biluo_tags_from_offsets(doc, entities, missing="O"):
biluo = ["-" for _ in doc]
# Handle entity cases
for start_char, end_char, label in entities:
if not label:
if start_char in starts:
biluo[starts[start_char]] = "O"
else:
for token_index in range(start_char, end_char):
if token_index in tokens_in_ents.keys():
raise ValueError(
@ -127,7 +131,7 @@ def biluo_tags_from_offsets(doc, entities, missing="O"):
break
else:
biluo[token.i] = missing
if "-" in biluo:
if "-" in biluo and missing != "-":
ent_str = str(entities)
warnings.warn(
Warnings.W030.format(